The Organic Chemistry of Drug Synthesis

THE ORGANIC OF DRUG CHEMISTRY SYNTHESIS VOLUME 4 DANIEL LEDNICER National Cancer Institute Bethesda, Maryland LESTER...

Author: Dr Daniel Lednicer

528 downloads 1703 Views 2MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

THE ORGANIC OF DRUG

CHEMISTRY

SYNTHESIS

VOLUME 4

DANIEL LEDNICER National Cancer Institute Bethesda, Maryland LESTER A. MITSCHER Department of Medicinal Chemistry The University of Kansas Lawrence, Kansas with GUNDA I. GEORG Department of Medicinal Chemistry The University of Kansas Lawrence, Kansas

A Wiley-Interscience Publication John Wiley & Sons, Inc. New York / Chichester / Brisbane / Toronto / Singapore

Copyright © 1990 by John Wiley & Sons, Inc. All rights reserved. Published simultaneously in Canada. Reproduction or translation of any part of this work beyond that permitted by Section 107 or 108 of the 1976 United States Copyright Act without the permission of the copyright owner is unlawful. Requests for permission or further information should be addressed to the Permissions Department, John Wiley & Sons, Inc. Library of Congress Cataloging in Publication Data: (Revised for volume 4) Lednicer, Daniel, 1929The organic chemistry of drug synthesis. "A Wiley-Interscience publication." Includes bibliographical references and Index. 1. Chemistry, Pharmaceutical. 2. Drugs. 3. Chemistry, Organic-Synthesis. I. Mitscher, Lester A., joint author. II. Title. [DNLM 1. Chemistry, Organic. 2. Chemistry, Pharmaceutical. 3. Drugs-Chemical synthesis. QV 744 L473o 1977] RS403.L38 615M9 76-28387 ISBN 0-471-85548-0(v. 4) Printed in the United States of America 10 9 8 7 6 5 4 3 2 1

We dedicate this book to Beryle and Betty who continue to support us in every imaginable way and to the memory of Katrina Mitscher-Chapman (1958-1987) who was looking forward with her customary enthusiasm to helping us prepare the manuscript.

I cannot tell how the truth may be; I say the tale as 'twas said to me.

Sir Walter Scott, "The Lay of the Last Minstrel"

Preface

Over a decade and a half have flown by since we started on the preparation of the first volume in this series. We did not at that time envisage a series at all but simply a book which filled what we then perceived as a vacuum. There were not in print in the midnineteen seventies any contemporary monographs in the English language dedicated to the synthesis of medicinal agents. The result was the original Organic Chemistry of Drug Synthesis. The reception accorded that volume confirmed that there was indeed a place for a book devoted to that subject matter. Having laid the groundwork, it seemed worthwhile to rectify a number of omissions present in the book and at the same time to bring the coverage for compounds included in the compilation to a common date. The result was of course Volume 2 and the birth of a series. The next volume, 3, was produced at the time we again felt the need to update our narrative; a semidecenial period was settled upon since it seemed to represent the best compromise between currency and a sufficient body of material to merit treatment in a monograph. The volume at hand continues the series; it covers the chemistry of those compounds which have been granted a United States Adopted Name (USAN) in the five years between 1983 and 1987. The bulk of the references thus fall in the 1980s; the reader will note occasional much older references. We suppose that those represent compounds which were synthesized many years ago and set on the shelf at that time; they were then revived for clinical development for one reason or another and a USAN applied for. It is well known that regulatory approval of new chemical entities has slowed markedly over the past decade. Some would even argue that the very rate of decrease is accelerating. This phenomenon has been attributed to a wide variety of causes, none of which are particularly germane to this volume. It is thus surprising, and pleasing, to note that the decreased probability of bringing a given new chemical entity to market has not led to a diminution in the rate of acquisition of new generic names as noted in USAN and USP Dictionary of Drug Names. The 300 odd compounds discussed in this volume are within a few entities of the number covered in the preceding volume. The acquisition of 60 new generic names each year has been so uniform over the past decade that this should perhaps be recognized as a new physical constant! This relatively steady rate of addition of new generic names has resulted in books which are quite uniform in size, at least after accounting for the text which was used to bring the subject up to date . The individual chapter titles do not show a corresponding uniformity; the composition of

x

PREFACE

the more recent volumes in some ways represents a socio economic history of research in medicinal chemistry. The first volume in this series, for example, contained a sizable chapter devoted to compounds based on the phenothiazine nucleus. This had disappeared by the second volume due to a dearth of new material. This in all probability simply represents a shift away from the research which took place on these compounds in the midnineteen fifties. Occasional chapters have lasted through all four volumes. One of these, to the authors' surprise is that devoted to " Steroids." That particular chapter is, however, by now a mere shadow of those which appeared in the first two volumes. Some chapters have persisted but changed significantly in content. "Alicyclic Compounds" has evolved from a collection of miscellany to a virtual compendium of prostaglandin syntheses. The diligent reader will note that succeeding volumes increasingly show agents which are the result of rational drug design of the synthesis targets. The older rationale for preparing specific compounds—to produce a hopefully superior and clearly patentable modification of a successful new drug—still however persists. Note that the present volume lists seven quinolone antibacterial agents, the same number of dihydropyridine calcium channel blockers, and no fewer than an even dozen angiotensin-converting enzyme inhibitors. Once the initial lead is discovered, a very significant expenditure of effort takes place; this persists until it becomes clear that no further improvements are taking place and that new entries are unlikely to gain a share of the market, This book is addressed primarily to practitioners in the field who seek a quick overview of the synthetic routes which have been used to access specific classes of therapeutic agents. Publications of syntheses of such compounds in the open literature remains a sometimes thing. One can, however, be certain that any compound which has commercial potential will be covered by a patent application. Many of the references are thus to the patent literature. Graduate students in medicinal and organic chemistry may find this book useful as an adjunct to the more traditional texts in that it provides many examples of actual applications of the chemistry which is the subject of their study. This volume, like those which came before, presumes a good working knowledge of chemical synthesis and at least nodding acquaintance with biology and pharmacology. Finally, the authors express their gratitude to Ms. Vicki Welch who patiently and skillfully prepared the many versions of this book including the final camera ready copy. Rockville, Maryland Lawrence, Kansas Lawrence, Kansas January, 1990

DANIEL LEDNICER LESTER A. MITSCHER GUNDA I. GEORG

Contents Chapter 1.

Chapter 2.

Chapter 3.

Chapter 4.

Chapter 5.

Aliphatic and Alicyclic Compounds 1, Acyclic Compounds 2. Alicyclic Compounds 3. Prostaglandins 4. Organoplatinum Complexes References Monocyclic Aromatic Compounds 1. Phenylpropanolamines 2. Phenoxypropanolamines 3. Alkylbenzenes and Alkoxybenzenes 4. Derivatives of Aniline 5. Benzoic Acid Derivatives 6. Diphenylmethanes 7. Miscellaneous Compounds References Polycyclic Aromatic Compounds and Thier Reduction Products L Naphthalenes and Tetralins 2. Indanes and Indenes 3. Fluorenes 4. Anthraquinones 5. Reduced Anthracenes References Steroids 1. Estranes 2. Androstanes 3. Pregnanes References Five-Membered Ring Heterocycles 1. One Heteroatom 2. Two Heteroatoms 3. Three Heteroatoms References

1 1 4 8 15 17 19 19 25 29 35 39 46 49 52 55 55 58 62 62 63 64 65 65 68 70 77 79 79 85 98 98

CONTENTS Chapter 6.

Chapter 7.

Six-Membered Ring Heterocycles 1. Pyridines 2. Dihydropyridines 3. Pyrimidines 4. Piperazines 5. Miscellaneous Compounds References Five-Membered Ring Benzofused Heterocycles 1. Benzofuranes 2. Indolines 3. Benzothiaphenes 4. Benzisoxazoles 5. Benzoxazoles 6. Benzimidazoles 7. Benzothiazole References

Six-Membered Ring Benzofused Heterocycles 1. Chromones 2. Benzodioxanes 3. Quinolines and Carbostyrils 4. Quinolones 5. Tetrahydroisoquinolines 6. Benzazepines 7. Benzothiepins 8. Quinazolines and Quinazolinones 9. Phthalazines 10. Benzodiazepines References Bicyclic Fused Heterocycles Chapter 9. 1. Indolizines 2. Pyrrolizines 3. Cyclopentapyrroles 4. Imidazopyridines 5. Purinediones 6. Purines 7. Triazolopyrimidines 8. Triazolopyridazines 9. Pyrimidinopyrazines 10. Pyridazinodiazepines 11. Thiazolopyrimidones 12. Thienopyrimidines 13. Thienothiazines 14. Pyrazolodiazepinones References

Chapter 8.

101 101 106 112 118 120 123 125 125 128 129 130 131 131 134 135 137 137 138 139 141 146 146 148 148 151 153 153 157 157 157 158 161 165 165 168 168 169 169 171 172 173 174 174

CONTENTS

xm

Chapter 10.

111 111 181 182 193 197

Chapter 11.

199 199 200 201 201 202 203 203 205 205 205 206 208 208 209 210 210 211 212 212 213 213 214 215 215 217 217 218 219 219 220 221 223 231 249

p-Lactam Antibiotics 1. Penicillins 2. Carbapenems 3. Cephalosporins 4. Monobactams References

Miscellaneous Heterocycles 1. Phenothiazines 2. Benzocycloheptapyridines 3. Carbazoles 4. Dibenzazepines 5. Dibenzoxepines 6. Pyridobenzodiazepines 7. Benzopyranopyridines 8. Pyrroloisoquinolines 9. Pyrazoloquinolines 10. Naphthopyrans 11. Benzodipyrans 12. Furobenzopyrans 13. Pyranoquinolines 14. Dibenzopyrans 15. Benzopyranopyridines 16. Thiopyranobenzopyrans 17. Pyrazinopyridoindoles 18. Thienobenzodiazepines 19. Imidazoquinazolinones 20. Imidazopurines 21. Pyrazinoisoquinolines 22, Pyrazinopyrrolobenzodiazepines 23. Imidazoquinolines 24. Oxazoloquinolines 25. Thiazolobenzimidazoles 26. Pyrimidoindoles 27. Ethenopyrrolocyclobutisoindoles 28. Thienotriazolodiazepines 29. Imidazobenzodiazepines 30. Imidazobenzothiadiazepines References Cross Index of Drug! Cumulative Index, Vols. 1-4 Index

T H E ORGANIC OF D R U G VOLUME 4

CHEMISTRY

SYNTHESIS

1

Aliphatic and

Alicyclic C o m p o u n d s

1. ACYCLIC COMPOUNDS There are relatively few important drugs which are alicyclic. Other than inhalation anesthetics, which are a special case, the compounds in the acyclic aliphatic class owe their activity to the functionality present and its specific spacing on the aliphatic framework. Thus, in most instances, the framework itself is not of comparable importance to the functionality attached to it. Caracemide (3) is an antitumor agent. This simple molecule is constructed by reacting acetohydroxamic acid (1) with methylisocyanate (2) promoted by triethylamine. The resulting O,N-biscarbamate (3), caracemide, is metabolized readily either by deacetylation or by decarbamoylation and its antitumor properties are believed to result from the reactivity of the resulting metabolites with DNA [1]. MeCONHOH (1)

+

2 MeN=C = O (2)

—*~

MeCONOCONHMe CONHMe (3)

Viral infections continue to be significant causes of morbidity and mortality and at the same time continue to be resistant to treatment by small molecules. Avridine (6) is an antiviral compound which has shown some activity in a variety of animal tests apparently based upon its ability to stimulate a number of cells to produce the high molecular weight endogenous antiviral substance interferon. Thus, the compound is believed to operate indirectly by stimulating the body's own natural defenses against viral penetration into host cells. Avridine is synthesized by 1

2

Aliphatic and Alicyclic Compounds

alkylating N-(3-aminopropyl)diethanolamine (5) with octadecyl bromide (4) using potassium carbonate in the usual fashion [2]. Me(CH2)16CH2 v Me(CH2)nBr

+ H2N(CH2)3N(CH2CH2OH)2

-

CH2CH2OH N(CH2)3N

Me(CH2)16CH2 ' (4)

(5)

CH2CH2OH (6)

Much attention has been focused upon the exciting promise of enzyme activated enzyme inhibitors for potential use in therapy. In contrast to the ordinary alkylating agents which are aggressive chemicals in the ground state and, thus, lack specificity in the body and produce many side effects and unwanted toxic actions, the so-called K-cat inhibitors or suicide substrates turn the enzyme's catalytic action against itself. The enzyme first accepts the suicide substrate as though it were the normal substrate and begins to process it at its active site. At this point, it receives a nasty surprise. This intermediate now is not a normal substrate which peacefully undergoes catalytic processing and makes way for another molecule of substrate, but rather is an aggressive compound which attacks the active site itself and inactivates the enzyme. As the suicide substrate is only highly reactive when processed by the enzyme, it achieves specificity through use of the selective recognition features of the enzyme itself and it works out its aggression at the point of generation sparing more distant nucleophiles. Thus, much greater specificity is expected from such agents than from electrophiles which are highly reactive in the ground state. Eflornithine (10) represents such a suicide substrate. Cellular polyamines are widely held to be involved in cellular growth regulation and, in particular, their concentration is needed for accelerated growth of neoplastic cells. The enzyme ornithine decarboxylase catalyzes a rate determining step in cellular polyamine biosynthesis and a good inhibitor ought to have antitumor activity. The synthesis of eflornithine starts with esterification of the amino acid ornithine (7) followed by acid-catalyzed protection of the two primary amino groups as their benzylidine derivatives (8). The acidic proton is abstracted with lithium diisopropylamide and then alkylated with chlorodifluoromethane to give 9. This last is deprotected by acid hydrolysis to give eflornithine (10) [3].

Aliphatic and Alicyclic Compounds Ornithine decarboxylase is a pyridoxal dependent enzyme. In its catalytic cycle, it normally converts ornithine (7) to putrisine by decarboxylation. If it starts the process with eflornithine instead, the key imine anion (11) produced by decarboxylation can either alkylate the enzyme directly by displacement of either fluorine atom or it can eject a fluorine atom to produce vinylogue 12 which can alkylate the enzyme by conjugate addition. In either case, 13 results in which the active site of the enzyme is alkylated and unable to continue processing substrate. The net result is a downturn in the synthesis of cellular polyamine production and a decrease in growth rate. Eflornithine is described as being useful in the treatment of benign prostatic hyperplasia, as an antiprotozoal or an antineoplastic substance [3,4]. CHF2

H2N(CH2)3CHCO2H

I ==N(CH2)3CCO2Me

C6H5CH ==N(CH2)3CHCO2Me

NH2

N==CHC 6 H 5

(7)

(8) (9)

CHF 2

jCHF

„ CHF-Enz

H2N(CH2)3CCO2H NH2

(10)

+

^CHPy

(11)

+

^CHPy

(12)

+^CH

(13)

Py = pyridoxal phosphate O n e interesting metabolic theory is that glucose and lipid levels in the blood affect each other's metabolism. G l u c o s e metabolism is disturbed in sugar diabetes and s o m e of the toxic effects of the resulting metabolic imbalance is believed to be due to enhanced oxidation of fatty acids as an alternate food. It is theorized that inhibitors of fatty acid oxidation could reverse the cycle in favor of glucose utilization. S o d i u m palmoxirate (19) was selected as a potential oral antidiabetic agent of a n e w type based upon this premise. Its synthesis begins by alkylating

4


methyl malonate with tridecylbromide (14) to give 15 and partially hydrolyzing the product to monoester 16. Next, treating the monomethylester with diethylamine and aqueous formaldehyde gives the desired alkyl acrylate ester 17. This is epoxidized with m-chloroperbenzoic acid and the resulting glycidic ester (18) is carefully hydrolyzed to give palmoxiric acid as its water soluble sodium salt (19). Palmoxirate is a potent hypoglycemic agent following oral administration to several animal species [5]. Me(CH2)i3Br

(14)

^Me(CH2)13CHCO2Me I CO2R (15);R«Me (16); R « H

2. ALICYCLIC COMPOUNDS An interesting appetite suppressant very distantly related to hexahydroamphetamines is somanladine (24). The reported synthesis starts with conversion of 1-adamantanecarboxylic acid (20) via the usual steps to the ester, reduction to the alcohol, transformation to the bromide (21), conversion of the latter to a Grignard reagent with magnesium metal, and transformation to tertiary alcohol 22 by reaction with acetone. Displacement to the formarnide (23) and hydrolysis to the tertiary amine (24) completes the preparation of somantadine [6],

(20); R = CO2H (21); R = CH2Br

(22); X = OH (23); X = NHCHO (24); X » NH2

Brain tumors are hard to treat in part because many antitumor agents which might otherwise be expected to have useful activity are too polar to pass the blood brain barrier effectively and fail to reach the site of the cancer. Nitrogen mustards are alkylating agents which fall into the category of antitumor agents which do not penetrate into the CNS. It is well known that a number of hydantoins pass through the highly lipid capillary membranes and, indeed, a number of CNS


5

depressants possess this structural feature. Combination of a hydantoin moiety to serve as a carrier with a latentiated nitrogen mustard results in spiromustine (28). Spiromustine is metabolized in the CNS to the active moiety, bis(chloroethanamine) (29). The synthesis begins with 5,5pentamethylenehydantoin (25) which is alkylated to 26 by reaction with l-bromo-2-chloroethane. Reaction of 26 with diethanolamine promoted by in situ halogen exchange with sodium iodide (Finkelstein reaction) leads to tertiary amine 27. The synthesis is completed by reacting the primary alcoholic moieties of 27 with phosphorus oxychloride [7]. H N

C

H

o ^

NR

JP

£ ^ N ^ N ^

X

HN(CH2CH2C1)2

o

(25); R = H (26); R - CH2CH2C1

(27); X = OH (28); X = Cl

(29)

Some alicyclic 1,2-diamine derivatives have recently been shown to have interesting CNS properties. For example, eclanamine (34) is an antidepressant with a rapid onset of action. The reasons for its potency are not as yet clear but pharmacologists note that the drug desensitizes adrenergic alpha-2 receptors and antagonizes the actions of clonidine. The synthesis of eclanamine starts with attack of cyclopentene oxide (30) by dimethylamine (to give 31). This product is converted to the mesylate by reaction with sodium hydride followed by mesyl chloride. Attack of COEt

(30)

(31);X = OH (32); X = OSO2CH3 (33); X = NH

OO (35)

(34)

6


the product (32) by 3,4-dichloroaniline leads to trans-diamine 33. The stereochemical outcome represents a double rear side displacement. The synthesis is completed by acylation with propionic anhydride to give eclanamine (34) [8], A chemically related agent, bromadoline (36) is prepared by an analogous series of reactions starting with cyclohexene oxide (35). Bromadoline is classified as an analgesic [9]. A structurally unrelated agent is tazadolene (40). The synthesis of tazadolene begins with p-keto ester 37 and subsequent enamine formation with 3-amino-l-propanol followed by hydrogenolysis to give 38. This phenylhydroxymethyl compound is then dehydrated with hydrochloride acid to form olefin 39. Treatment with bromine and triphenylphosphine effects cyclization to form the azetidine ring of tazadolene [10]. OH

cc — o (37)

,-CHPh .N(CH2)3OH H (38)

CHPh

H (39) (40) C e r t a x a e t ( 4 4 ) i s a p r o d r a g o f r t a n e x a m c i a c d i . T h e a l t e r i s a in ti n ighab istrtsion h t e a c v t i a o t i n o f p a l s m n i o g e n t o p a l s m n i . T h e r e s u t l i s t o p r e v i t e s n t i a l u c l e r s . P r o d r u g s a r e o f v a u l e a s h t e y a o l w g r e a e t r a b s o th rtaeotinesteb y s u p p r e s n i g , n i t h s i c a s e , h t e a m p h o e c t r i n a u t r e o f h t e d r u g . T r c f i a o t i n o f 3 £ ( h y d r o x y p h e n y p ) l r o p o i n c i a c d i ( 4 2 ) b y r t a n s 4 c y a n o c y c o l h e x a n e c can hd olrd ieRa(n 4ey1).nT h e p r o d u c t ( 4 3 ) i s r e d u c e d t o c e t r a x a t e b y c a t a y l t c i h y d r o cikel [11].


COC1

CH2CH2CO2H r r

c

o

n

(43); R - CN (44); R = CH2NH2 Among the most successful drugs of recent years have been the group of antihypertensive agents which act by inhibition of the important enzyme, angiotensin-converting enzyme (ACE). The renin-angiotensin-aldosterone system exerts an important control over blood pressure and renal function. One of the key steps in the process is the conversion of angiotensinogen to angiotensin I by the enzyme renin. Angiotensin I, an octapeptide (Asp-Arg-Val- Tyr-Ile-His-Pro-PheHis-Leu), is cleaved of two amino acids by ACE to a hexapeptide, angiotensin II (Asp-Arg-ValTry-Ile-His- Pro-Phe), a powerful pressor hormone. The majority of the inhibitors of this important enzyme are treated in a later chapter. One of the structurally more interesting representatives, however, is pivopril (50), an orally active prodrug with a masked sulfhydryl group (protected by a pivaloyl ester moiety) and, instead of possessing the usual chiral C-terminal proline residue, has an achiral N-cyclopentylglycine moiety. The synthesis begins with the reaction of the t-butylester of N-cyclopentyl glycine (45) with (S)-3-acetylthio-2-methylpropionyl chloride (46) to give amide 47. The acetyl group is selectively cleaved with ammonia in methanol to give 48. The thiol group is reprotected by reaction with pivaloyl chloride to give 49 and the carboxyl protecting group is removed by selective reaction with trimethylsilyl iodide to give pivopril (50) [12]. The structural relationship of pivopril to the commercially important analogues captopril (51) and enalaprilat (52) is readily apparent. Retinoids are needed for cellular differentiation and skin growth. Some retinoids even exert a prophylactic effect on preneoplastic and malignant skin lesions. Fenretinide (54) is somewhat more selective and less toxic than retinyl acetate (vitamin A acetate) for this purpose. It is synthesized by reaction of all trans-retinoic acid (53), via its acid chloride, with rj-aminophenol to give ester 54 [13].

Ap ilhacti and Acilyccil Compounds NHCH C C 2O 2M 3e C 1 (45) (46C )O

CC

64(7R = -AcH ((4498;)));; R R -M C CO 3e

M C OSCHj^ C O N C H C H 3e 2O 2 £O H rvef O tH 2 2 (50) (51) (52) C O R

((5534));; R = O H R - NH3. PROSTAGLAND N IS The prosatgalndnis connitue theri stateyl progres otwards cn ilcial use. Theri pr regualtors are wel estabsih led but theri use for oh ter h terapeucit neds is co weah tl of sd ie efects. Theri use as cytoprotecvtie agenst and anstiecretory age oloks promn sig,however,and there is osme hope for h te calsscial agenst svie agents; oh terwsei much of the curent excetiment wh ti these compounds control the boisynthesi of parctiualr prosatnodis or to moduaelt theri acotin at Motsn iterest centers around the oh ter producst of h te arachdionci acd i cascad boxanes and elucotreines where n itervenotin promseis controlof dsiorders of hem


9

inflammation. Few of these substances have progressed far enough to be the subject of paragraphs in this work as yet. Alfaprostol (55) is a luteolytic agent used injectably for scheduling of estrus in mares for purposes of planned breeding. It is also used for treatment of postweaning anestrus in economically important farm animals. For these purposes, alfaprostol is more potent than naturally occurring prostaglandin F2-alpha. Notable molecular features of the alfaprostol molecule are the acetylenic linkage at C-13, the methyl ester moiety (which is rapidly removed in vivo) and the terminal cyclohexyl moiety which inhibits some forms of metabolic inactivation. The synthesis begins with lactol 56 which undergoes Wittig reaction with methyl 5-triphenylphosphoniumvalerate (57) using dimsyl sodium as base. Dehalogenation occurs concomitantly to produce partially protected condensation product 58. Deblocking to alfaprostol is brought about by oxalic acid [14].

Othp

Othp (56)

(57)

(58); R = thp (55); R - H

Another luteolytic agent, fenprostalene (62) contains an alleneic linkage in the upper sidechain and terminates in a phenoxy moiety in the lower. Its synthesis begins with lactol 59 (presumably the product of a Wittig olefination of the Corey lactol and suitable functional group manipulation). Lactol 59 is reacted with lithio 4-carbomethoxybut-l-yne and the resulting secondary carbinol acetylated with acetic anhydride to give substituted acetylene 60. The allene moiety (61) is produced by reaction with copper (II) bromide and methyl lithium. The tetrahydropyranyl ether protecting groups are then removed by treatment with acetic acid, the ester groups are hydrolyzed with potassium carbonate, and the carboxy group is reprotected by diazomethane methylation to give fenprostalene [15].

10


OH

AcO AoO

i Othp Othp (59)

(60)

A prostaglandin closely related to fenprostalene is enprostil (63). Enprostil belongs to the prostaglandin E family and is orally active in humans in reducing gastric acid and pepsin concentration as well as output. It is effective in healing gastric ulcers in microgram doses and is under consideration as an antisecretory, antiulcerative agent. The synthesis begins with intermediate 61 by removing the protecting THP ether groups with acetic acid (64) andthen replacing them with t-butyldimethylsilyl groups by reaction with t-butyldimethylsilyl chloride and imidazole. This is followed by hydrolysis of the ester moieties with potassium carbonate and reesterification of the carboxy moiety with diazomethane to produce intermediate 65. The solitary free alcoholic hydroxyl at C-9 is oxidized with Collins' reagent and the silylether groups are removed with acetic acid to give enprostil (63) [15].

(64) R=Ac; Y=H (65)R=H; Y=SiMe2i-Bu


11

CO2Me

(63) Enisoprost (70) is an antiulcerative/cytoprotective prostaglandin. In addition to the wellknown property of E series prostaglandins to inhibit gastric secretion of HC1 and pepsin, these agents enhance ulcer healing by stimulating formation of the mucin protective layer over the stomach lining. The well-known ulcer promoting action of nonsteroidal antiinflammatory agents such as aspirin can be rationalized by invoking the reversal of this effect. Thus, useful antiulcer properties can be anticipated at very low doses of certain prostaglandins (offsetting their cost) and this has been confirmed in the clinic. One of the side effects of such prostaglandins which must be minimized is diarrhea and cramps. In the enisoprost molecule this has been accomplished by moving the C-15 OH group of ordinary prostaglandins to C-16. This is consistent with antiulcer activity but reduces other side effects. Presumably these results reflect different structural needs of the different receptors. The addition of the methyl group at C-16 prevents oxidative inactivation of the molecule which would involve ketone formation at C-16 otherwise. This devise is a common stratagem used previously, for example, with methyltestosterone. The presence of a cis double bond at C-4 is also known to inhibit oxidation beta to the carboxyl group. Thus enisoprost carries a number of interesting design features. The synthesis concludes by conjugate addition of mixed cuprate 68 to unsaturated ketone 69. The product, enisoprost, is the more stable isomer with the two new side chains trans. The mixed cuprate is made from protected acetylene alcohol 66 by photosensitized trans addition of tri-n-butyltin hydride to give organostannane 67. Successive transmetalations with butyl lithium and then copper 1-pentyne leads to the necessary mixed cuprate (68) for the above sequence [16], Gemeprost (73; 16,16-dimethyl-trans-A2-prostaglandin-E1) is dramatically more potent on a dosage basis as an abortifacient than prostaglandin E2 itself and has fewer side effects. The gem-dimethyl groups at C-16 protect the alcohol moiety at C-15 from rapid metabolic oxidation.

12


(66) > NsX i (94)

n^N

2

UJt-12

(95)

2_

H N

"

(96)

Spiroplatin (99) in animal studies showed excellent antileukemic activity and was less


17

nephrotoxic than cisplatin. Unfortunately when it was tried clinically, little antitumor activity could be demonstrated and it was hard to determine safe doses to use in humans so it was ultimately dropped. Its synthesis starts with potassium tetrachloroplatinate (91) which is reacted with spiro-l,3-propanediamine 97 and potassium iodide to give complex platinate 98. This is treated with silver sulfate to produce spiroplatin [24].

V _ A _ NH2

\

(97)

(98)

(99)

The only prominent antitumor tetravalent platinum complex so far is iproplatin (102). In vitro it has been shown to cause interstrand DNA-breaking and cross linking. Free radical scavengers inhibit these effects. The complex is less neurotoxic and less nephrotoxic than cisplatin. Its synthesis begins with hydrogen peroxide oxidation of cis-dichlorobis(isopropylamine) platinum (100) to the dimethylacetamide complex 101. The latter is heated in vacuum to liberate iproplatin [25].

Me2CHNH2 ^

y

Pt

Me2CHM-I2

Cl

Cl (100)

Me2CHNH2

OH

-MeCONMe2 t Me2CHNH2' I XC1 OH (101)

OH Me2CHNH2 ^ | c , Me2CHNH2' | ^ Cl OH (102)

REFERENCES 1. W. Reifschneider, Ger, Offen., 3,305,107 (1983) via Chem. Abstr., 100: 6,101p (1984). 2. T. H. Cronin, H. Faubl, W. W. Hoffman, and J. J. Korst, U. S. Patent 4,034,040 (1977) via Chem. Abstr., 87: 134,479t (1977). 3. P. Bey and M. Jung, U. S. Patent, 4,330,559 (1982) via Chem. Abstr.t 97: 144,373z (1982). 4. B. W. Metcalf, P. Bey, C. Danzin, M. L. Jung, P. Casara, and J. P. Vevert, /. Am. Chem. Soc, 100,2551(1978). 5. W. Ho, G. F. Tutwiler, S. C. Cottrell, D. J. Morgans, O. Tarhan, and R. J. Mohrbacher, /.

18

Aliphatic and Alicyclic Compounds Med. Chem., 29, 2184 (1986).

6. B. V. Shetty, U. S. Patent, 4,100,170 (1978) via Chem. Abstr., 90: 86,875g (1978). 7. G. W. Peng, V. E. Marquez, and J. S. Driscoll, J. Med. Chem., 18, 846 (1975). 8. J. Szmuszkovicz, U. S. Patent, 4,159,340 (1979) via Chem. Abstr., 91: 157,342q (1979) and U. S. Patent, 4,156,015 (1979) via Chem. Abstr., 91: 74,259s (1979). 9. J. Szmuszkovicz, U. S. Patent, 4,215,114 (1980) via Chem. Abstr., 94: 103,032g (1980). 10. J. Szmuszkovicz, Eur. Pat. Appl., 85,811 (1983) via Chem. Abstr., 100: 6,31 lg (1983). 11. Anon., Japan Kokai, 59/134,758 (1984) via Chem. Abstr., 101: 230,035y (1984) and C. M. Svahn, F. Merenyi, and L. Karlson, J. Med. Chem., 29, 448 (1986). 12. J. T. Suh, J. W. Skiles, B. E. Williams, R. D. Youssefyeh, H. Jones, B. Loev, E. S. Neiss, A. Schwab, W. S. Mann, A. Khandwala, P. S. Wolf, and I. Weinryb, /. Med. Chem., 28, 57 (1985). 13. Y. F. Shealy, J. L. Frye, C. A. O'Dell, M. C Thorpe, M. C. Kirk, W. C. Coburn, Jr., and M. B. Sporn, /. Pharm. Sci., 73,745 (1984). 14. C. Gandolfi, R. Pellegata, R. Caserani, and M. M. Usardi, U. S. Patent 4,035,415 via Chem. Ator.,85:7,748n(1976). 15. J. M. Muchowski and J. H. Fried, U. S. Patent, 3,985,791 (1979) via Chem. Abstr., 92: 146,339p(1980). 16. P. W. Collins, E. Z. Dajani, R. Pappo, A. F. Gasiecki, R. G. Bianchi, and E. M. Woods, J. Med. Chem., 26, 786 (1983). 17. H. Suga, Y. Konishi, H. Wakatsuka, H. Miyake, S. Kori, and M. Hayashi, Prostaglandins, 15, 907 (1978). 18. H. C. Kluender, W. D. Woessner, and W. G. Biddlecom, U. S. Patent, 4,132,738 (1979) via Chem. Abstr., 90: 151,668h (1979). 19. J. E. Bimbaum, P. Cervoni, P. S. Chan, S. M. Chen, M. B. Floyd, C. V. Grudzinskas, and M. J. Weiss, /. Med. Chem., 25, 492 (1982). 20. P. A. Aristoff, P. D. Johnson, and A. W. Harrison, J. Org. Chem., 48, 5341 (1983). 21. B. Rosenberg, L. Van Camp, J. E. Trosko, and V. H. Mansour, Nature, 111, 385 (1969). 22. G. B. Kauffman and D. O. Cowan, Inorg. Syn., 7, 239 (1963). 23. R. C. Harrison, C. A. McAuliffe, and A. M. Zaki, Inorg. Chem. Acta, 46, L15 (1980). 24. J. Berg, F. Verbeek, and E. J. Bluten, U. S. Patent, 4,410,544 (1983) via Chem. Abstr., 100: 29,622y (1983). 25. P. C. Hydes and D. R. Hepburn, Belg. 890,209 (1982) via Chem. Abstr., 97: 61,004d (1982).

2

Monocyclic Aromatic

Compounds

Benzene rings constitute quiterigid,flat, relatively lipophilic moieties with considerable electron density. Groups attached to a benzene ring not only modulate these properties by their relative electron donating or withdrawing character, but also occupy well-defined spatial positions by virtue of the bond angles which form those links. These properties of the aromatic ring enhance uniqueness and fit to receptor sites for endogenous mediators. The benzene ring thus forms the nucleus for a number of pharmacophores. 1. PHENYLPROPANOLAMINES The adrenergic nervous system plays a key role in the regulation of the cardiovascular system. Functions such as performance of the heart muscle and blood pressure are directly affected by levels of the chemical transmitters of the adrenergic system, epinephrine (1) and norepinephrine (2). Drugs which act on the cardiovascular system by interacting with the adrenergic system have had a major impact on treatment of cardiovascular diseases. These agents range from compounds which act as antagonists at the receptors for beta adrenergic agents (beta blockers) to receptor agonists used to increase contractile force. Effects of epinephrine and norepinephrine result from interaction of those compounds with at least four adrenergic receptors: the alpha 1 and 2 receptors and the beta 1 and 2 receptors. Some of the side effects due to beta blockers such as the slowing of heart rate can be counteracted by administration of drugs which antagonize the alpha adrenergic receptors. The 19

20

Monocyclic Aromatic Compounds

antihypertensive agent labetalol (3) in fact includes both actions in a single molecule. The presence of two chiral centers in that molecule allows for the existence of two diastereomeric pairs. Preparation and testing of the individual optical isomers showed that each of these had a somewhat different combination of activities. A different conclusion would have been surprising in view of the fact that receptors themselves are made up of chiral molecules. In the event, it was ascertained that the R,R isomer exhibited the best combination of activities. The synthesis starts by condensation of readily available optically active (R)-(+)-alphamethylbenzylamine with 4-phenyl-2-butanone to form an imine which is itself reduced by hydrogenolysis (Raney nickel) to give a 9:1 mixture of the (R,R)-amine and the (R,S)-amine (4). This product (4) is then separated into its diastereomers by recrystallization of the corresponding hydrochlorides. Since the amine (4) was found to be inert to alkylation with phenacylhalides such as 7, it was debenzylated by hydrogenolysis (Pd/C) to give the primary (R)-amine 5. Reductive alkylation with benzaldehyde and hydrogen resulted in the formation of the N-benzyl derivative 6. Alkylation of the secondary amine 6 with bromoketone 7, followed by reduction (Pd/Pt/hydrogen) of the ketone group in 8 gives the alcohol 9. The relative remoteness of the ketone linkage from the chiral center leads to the formation of both diastereomers in a 1:1 mixture. The resulting diastereomers are separated by fractional crystallization. Removal of the benzyl substituents by means of catalytic reduction affords the secondary amine 10. There is thus obtained the optically active antihypertensive agent dilevalol (10) [1]. It has by now been well established that Parkinson's disease involves a deficiency of dopamine (11) in the brain. It has been further shown that any one of several stratagems for increasing levels of that neurotransmitter near the appropriate receptors will alleviate the symptoms of that disease. For example, a well-absorbed dopamine agonist which reaches the brain should thus be useful in treating that syndrome. Though ciladopa (16) at first sight closely resembles a beta blocker it should be noted that the presence of the hydroxyl group apparently does not interfere with dopaminergic activity. In addition, the compound lacks the secondary amino group which is thought to be indispensable for interaction with beta adrenergic receptors.


21

(2);R = Me

Me CM — NHCHCH C C 2H 2H 65 e H (4)

?+ M •rC —NH C eC H C — 2 2H H

R,R R,S .CONH2 -OCH2C6H5

O

II 2 BrCH^265

2~~ CHCHCH CH C H N 65 2

(7)

RHN— C —CH 2 CH—

-OCH 2 C 6 H 5

(8)

Me

'CONH?

C—CH 2 CH 2 C 6 H 5

OCHC 2H 65

CH 2 Ctt OH (9);(R,R

+R.S)

H

- HC H2NOC

(5); R = H (6);R = CHC 2H 65 O H C -HCH2NHCHCH C C 2H 2H 65 Me (3); (R,R +R,S + S,S (10); (R,R)


22

Condensation of piperazine with 2-methoxytropone gives the addition-elimination product 12 [2], Alkylation of the remaining secondary amino group with bromoketone 13, itself the product from acylation of dimethyl catechol, gives aminoketone 14. Reduction of the carbonyl group with sodium borohydride leads to secondary alcohols 15 and 16. Resolution of these two enantiomers was achieved by recrystallization of their tartrate salts to give ciladopa (16) [3].

- CH2CH2NH2 HO (11)

P -OMe +

HN

NH

-N

NH

(12)

p

OH r ~^NCFI CH-— OMe OMe

(IS); (5) (16); (*) Condensation of adipic acid derivative 17 with phenylethylamine in the presence of carbonyldiimidazole affords the bis-adipic acid amide 18. The synthesis is completed by reduction of the carbonyl groups with diborane followed by demethylation of the aromatic methoxy groups with hydrogen bromide the afford dopexamine (19) [3].


23

MeO—^_y~(CH2)2NHCO(CH2)4CO2H + H2NCH2CH2(17)

MeO—/_y-CH2CH2NHCO(CH2)4CONHCH2CH2MeO

CH2CH2NH(CH2)6NHCH2CH2—/_^ (19) Interposition of an amide function in the norepinephrine-like side chain in midodrine (25) affords a compound which retains a good measure of adrenergic activity. Acylation of dimethylhydroquinone with chloroacetyl chloride gives the chloroketone 20. The halogen is then converted to the amine 21 by any of a set of standard schemes, and the ketone reduced to an alcohol with borohydride (22). Acylation of the amino group in this last intermediate with chloroacetyl chloride affords the amide 23. The halogen is then displaced by azide and the resulting product (24) reduced catalytically to the glycinamide, midodrine (25) [4]. A compound closely related to classical adrenergic agonists in which the para hydroxy function is however replaced by an amino group has been investigated for its activity as a growth promoter in domestic animals. Acylation of the aniline derivative 26 with chloracetyl chloride will afford acetophenone 27; the amino-ketone 28 is obtained on reaction with isopropylamine. Removal of the protecting group (29) followed by reduction of the ketone affords cimaterol (30) 15].


24 O CCH2C1

MeO

MeO

(20)

OH MeO l.CHCH2NH2

MeO (f

V MeO

MeO (21)

MeO

(22)

OH O I II XCHCH2NHCCH2X

MeO

MeO

OH

O

,CHCH 2 NHCCH 2 NH 2

MeO (25)

(23);X = C1 (24); X = N3

s

NHCOMe

NHCOMe

(26)

(27)

OH . CN

Me2CHNHCH2Cv s

(29); R = COMe (30); R = H

(28)

NHCOMe


25

2. PHENOXYPROPANOLAMINES Compounds which act as antagonists at the receptors for beta sympathetic transmitters (beta blockers) have gained very wide acceptance as antihypertensive agents. It was found subsequent to their introduction that there are two populations of beta receptors; the beta-1 receptors are richest in the cardiovascular system; whereas beta-2 receptors are mostly found in the bronchi. Lack of receptor-type specificity led to bronchial spasm in some asthmatic individuals on ingestion of the earlier beta blockers. Much of the work outlined below had as its goal the preparation of agents which showed selectivity for beta-1 receptors. It will be noted that the great majority of beta blockers consist of phenoxypropanolamines. Many of the agents incorporate substituents ortho to the side chain in response to the observation that such groups increase potency; this is thought to be due to the bulk of that substituent, encouraging productive conformations. The synthetic schemes for these agents as a rule culminate in the introduction of the aminoalcohol side chain. The first step often consists in reaction of the appropriate phenolate with epichlorohydrin to give a glycidic ether such as 32; reaction with a primary amine, usually isopropylamine or tert-butylamine, leads to the amino alcohol. Application of this scheme to o-cyclohexylphenol (31), leads to exaprolol (33) [6],

OH

o(31)

o(32)

(3o 3)

Alkylation of the monobenzyl ether of hydroquinone 34 with mesylate 35, gives ether 36. Hydrogenolytic removal of the benzyl group gives phenol 37. This affords cicloprolol (38) when subjected to the standard alkylation scheme 17]. In much the same vein, alkylation of rj-hydroxyphenylethanol 39, obtainable from the corresponding phenylacetic acid, with epichlorohydrin

26


gives ether 40. Aminolysis with isopropylamine produces 41 which was alkylated with cyclopropylmethyl bromide to give betaxolol (42) [8]. Aminolysis with t-butylamine of epoxide 43 gives the beta blocker cetamolol (44) [9].

MeSO3CH2CH2OCH2—N' Cl (140)

OMe CO2H

OMe L

RHN' (141); R = H (142); R - COCH3

OMe

r X CONH—/ ~

RHN Cl (143); R = COMe

Cl (145)

,(144); R == H H 2 N—/C^

CONH

(156)


44 Et OMe (150)

(151); (S)

(148); X = H (149);X = Br

MeO2C HO—/~\—NH 2 (158)

(161)

(159)

(160)

One of the early steps in an allergic reaction consists in the release of a series of endogenous compounds referred to as mediators from sensitized cells. The finding in the early 1960s that cromolyn sodium, still the only approved drug of this class, blunts this reaction has led to an intense search for additional examples. It is of interest that a relatively simple anthranilic acid derivative has shown mediator-release inhibiting activity. Reaction of 3,4-drmethoxybenzaldehyde (167) with isatoic anhydride 168 gives the condensation product 169, which, upon hydrolysis, affords tranilast (170) [43].


C O C C O H C O 2H 2H 2C 2H 2H ocN H (163) (162) C O H 2 C O H (165) 2 (166) M O e C H O M O e (169)

45

o.

o

(168)

M O e/ MO e z (170) A substu ited benzoci acd i serves as precursor for the nontrciycil a mol(175). Seelctvie saponcfiaotin of ester 171 aofrds h te haf-alcd i 172. cholrd ie derviedrofm thsi nietrmedaiet (173) wh ti ammoanigvies h te am d h te alst by means ofh tiluim u alm nium hydrdie gvies bpienamol(175) [ CO E O tC H O C H H N 2t E 2 2 C 2H 2 ( 1 7 5 ) (((111777234)));;; R = O H (171) R = C l R = NH 2


46

6. DffHENYLMETHANES The sulfur analogue of the Hauser ortho-substitution rearrangement provides access to an arylacetic NSAID. Reaction of the aminobenzophenone 176 with ethyl methylthioacetate and tert-butyl hypochlorite gives the intermediate 178. The reaction probably proceeds by way of formation of the S-chlorinated sulfonium derivative 177; displacement on sulfur will lead to the salt 178. Treatment with triethylamine leads initially to the betaine 179. Electrocyclic rearrangement of that transient intermediate leads, after rearomatization, to the homoanthranilic acid 180. Internal ester-amine interchange leads then to indolone 181 [45]. The thiomethyl group is then removed with Raney nickel. Saponification of intermediate 182 affords bromfenac (183) [46].

,CH2CO2Et

ra2 + cH3scH2co2Et — - [CH3SCH2CO2E7]

(178)

(177)

CO2Et S—Me

(180)

(181); R = SMe (182); R = H

(179)

(183)


47

Gamma-aminobutyric acid (GABA) ranks among the numerous brain neurotransmitters; it exerts inhibitory activity on certain pathways and abnormalities in levels of GABA or in its transport and/or metabolism have been implicated in various CNS diseases including convulsions. Attempts to treat such disorders by administration of exogenous GABA are rendered difficult by the compound's poor penetration of the blood brain barrier. A series of imines of GABA or its potential metabolic precursors with phenolic benzophenones are apparently well enough absorbed to show andepileptic and anticonvulsant activity. Acylation of jg-chlorophenol with o-chlorobenzoyl chloride gives, after Fries rearrangement, phenol 184; formation of the imine between this ketone and butylamine affords fengabine (185) [47]. In a similar vein, acylation of 2,4-dichlorophenol with 2-benzoyl chloride gives the benzophenone 186, after Fries rearrangement; acylation of r> fluorophenoi with the same acid chloride gives the rearranged benzophenone 188. Formation of the imines of these ketones with the amide of GABA gives respectively tolgabide (187) [48], and progabide (189) [49].

NCH2CH2CH2CH3

(184)

(185)

•NCH2CH2CH2CONH2

fCH2CH2CH2CONH2

48

Monocyclic Aromatic Compounds It has been shown that glycine amides of aminobenzophenones are readily converted to the

corresponding benzodiazepines in vivo. Peptides which terminate in such a moiety should thus serve as a benzodiazepine prodrug after hydrolysis by peptidases. One of the glycine residues in lorzafone (194)is presumably removed metabolically in this manner to give a benzodiazepine precursor which spontaneously cyclizes. Acylation of benzophenone 190 with the trityl protected dipeptide 191, as its acid chloride 192, affords the amide 193. Removal of the trityl protecting group with acid yields lorzafone (194) [50].

O o II II R— CCH2NHCCH2NHC(C6H5)3

( 1 9 0)

(191); R-OH (192); R-Cl Me O

O

I II II ,N— CCH2NHCCH2NHR

(193); R - C(C6H5)3 (194); R = H The sedation side effect commonly observed on administration of classical antihistaminic drugs has been attributed in part to the ease with which many of these compounds cross the blood brain barrier. There have been developed recently a series of agents, for example, terfenadine (198), which cause reduced sedation by virtue of decreased penetration into the CNS. This is achieved by making them more hydrophilic. Synthesis of a related compound, ebastine (197),


49

starts by alkylation of 4-hydroxypiperidine with butyrophenone 195. Alkylation of the alcohol 196 at the 4 position with benzhydryl bromide leads to ebastine (197) [51]. O II

+

C1CH2CH2CH2C

(195)

II H O — ^ I S f C H 2 C H 2 C H 2 C — ^ y — CMe3 (196)

CMe3

(197)

H 0

—

c

— o—
the intermediate inline undergoes lmine exchange and then reduction to achieve this result [211

CH2CH2CH2NMt (61) The lmidazohne derivative cibenzohne (64) is a class I antiarrhythmic agent which has undergone clinical trials in the United States with apparently satisfactory results It is synthesized by diphenylcyclopropanation of acrylonitnle by thermal carbene generation from diphenyldiazo methane (62) to give 1 cyano 2,2 diphenylc>clopropane (63) Reaction of this with ethylenedia mine tosylate completes the synthesis of ciben/ohne (64) [221

Phv (63)

»-> (64)

Napamezole (68) is a dihydroimidazole derivative with antidepressant activity probably as a result of its combined a 2 adrenergic receptor blocking and serotonin uptake blocking proper ties It can be synthesized by Wittig olefination of p-tetralone (65) with diethyl (cyanomethyl) phosphonate (66) and base to give nitnle 67 lmidazohne construction on the latter was smoothly

88

Five-Membered Ring Heterocycles

accomplished by reaction with ethylenediamine and trimethylaluminum. The product (68) is napamezole 123],

[I T ]

+

(65)

(EtO)2POCH2CN (66)

(67)

(68)

Another imidazoline derivative, lofexidine (71), has different pharmacological properties, being an antihypertensive. It is pharmacologically reminiscent of clonidine. As expected for a central adrenergic a-blocker, a major side effect in the clinic was orthosiatic hypertension. The compound has been marketed. Its synthesis begins with alkylation of 2-ehloropropionitrile with 2,6-dichlorophenol (69), KI, and base. The ether produced (70) is converted to lofexidine by reaction with ethylene diamine and catalytic carbon disulfide |24) Mcv JCN I O

OH — (69)

C i

IN

x r a (70)

(71)

An imidazole derivative which is also a hypotensive agent by virtue of adrenergic a-2receptor blockade is imiloxan (75). Its synthesis begins by conversion of 2~cyanomethyl-l,4benzodioxane (72) to its imincxkthylether with anhydrous HC1 in ethanol (73), Reaction of the latter with aminoacetaldehyde diethylacetal and subsequent acid treatment produces the imidazole ring (74). Alkylation of 74 with ethyl iodide mediated by sodium hydride completes the synthesis [25].

(72)

( X r f " _ ^ ^ c r OEt (73)

(74)


89

Mefenidil (78) is a cerebral vasodilator which may be of value in treating geriatric cerebral circulatory problems. It can be synthesized by reacting benzamidine (76) with biacetyl to produce the highly reactive methylene benzimidazole adduct 77. Reaction of the latter with sodium cyanide completes the synthesis [26].

H keto moiety) to give 90 This is reacted with hydrazine to produce the dihydropyndazinone ring of ima/odan (91) [30,31]

(87), R = H X=NH2 (88),R=Me XP=CN

" TX (89)

(90)

" (91)

Nafimidone (93), an anticonvulsant compound, also contains an lmidazole moiety It seems to have been discovered by accident during a search for anufungal agents Its synthesis is straightforward involving displacement with lmidazole of the activated chlonne atom of chloromethyl-p-naphthylketone (92) [32]

Five Membered Ring Heterocycles

91

HoCl (93)

(92)

We continue to document the point that the imidazole ring rjer se though common enough in drugs, is not a general pharmacophore by considenng the antithrombotic agent dazoxiben (96) Imidazole itself has some thromboxane B 2 inhibitory action but is dramatically less active than analogues such as da/oxiben Inhibition of this important enzyme in the arachidonic acid cascade results in a surplus of prostacychne and net inhibition of blood clotting One convenient synthesis starts with the O chlorocthylether of p hydroxybcnzamidc (94) and proceeds bv displacement with imidazole to give 95 Hydrolysis of the amide function completes the s>nthesis of dazoxiben

CONH,

u

CONH,

CO2H

O

u

Cl

(96)

T u b u l o z o l e ( 1 0 1 ) is a n a n t i n e o p l a s t i c a g e n t b y v i r t u e o f its ability to inhibit formation

microtubule

M i c r o t u b u l e s are a sort of cellular scaffolding a n d cellular r e p r o d u c t i o n is

without the ability to m a k e microtubules Interestingly, o n l y the cis i s o m e r is active

T h u s , I a , t u m o r cell g r o w t h is i n h i b i t e d b y

impossible tubulo/ole

T h e s t r u c t u r e o f t u b u l o z o l e is s u c h a s to s u g g e s t t h a t it

w a s originally prepared as a potential antifungal agent

O n e o f t h e s y n t h e s e s s t a r t s w i t h rj

acetanilide (97) w h i c h condenses with c o m p l e x methanesulfonate 9 8 in the presence of K acetone to produce acetanilide 99

thio

2

Deblocking by hydrolysis via 100 followed by reaction

C O

3

with

in

Five-Mernbered Ring Heterocycles

92

ethylchloroformate completes this synthesis [35]. The synthesis of requisite intermediate 98 is straightforward from 2,4-dichloroacetophenone.

OSO2Mc (97)

(99); X = COMc (100); X = H (101); X = CO2Et

(98)

Infections by fungi are becoming increasingly prominent as more individuals undergo immunesuppression following intensive corticoid therapy, organ transplantation, anticancer therapy, or infection by the dreaded AIDS virus. Consequently an intensified search for relatively nontoxic, broad spectrum, and orally active antifungal agents has been undertaken. Many appropriately substituted imidazole derivatives, known collectively as the 'conazoles, have been prepared and some have found clinical use. Zinoconazole (103) is a typically complex example of the second generation of such compounds but is notable for its oral activity and for being fungicidal at realistic doses. In common with the other 'conazoles, zinoconazole blocks the enzymic demethylation of lanosterol by a cytochrome P-450 enzyme thus, preventing the formation of ergosterol, an important cell membrane constituent of many pathogenic fungi. Zinoconazole seems also to damage membranes, directly thereby exerting a 'cidal rather than a 'static effect. It is made readily by hydrazoue formation between ketone 102 and 2,6-dichlorophenylhydrazone [36].

(102)

(103)


93

Fenticonazole (106), on the other hand, is used topically to combat a wide variety of dermatophytes and yeasts, particularly Candida albicans. It can be synthesized from 2,4-dichlorophenacyl chloride (104) by reduction with borohydride and subsequent displacement with imidazole to give 105. This last undergoes ether formation with p-thiolphenylbenzyl chloride mediated by NaH to produce fenticonazole (106) [37].

OH (105)

(104)

E n i l c o n a z o l e ( 1 0 7 ) h a s b e e n m a r k e t e d f o r a n t i f u n g a l u s e in p l a n t s a n d a n i m a l s . It c a n b e s y n t h e s i z e d in a v a r i e t y of w a y s i n c l u d i n g o n e c l o s e l y a n a l o g o u s to that u s e d for f e n t i c o n a z o l e e x c e p t t h a t t h e a l k y l a t i n g g r o u p is a l l y l c h l o r i d e 1 3 9 ] .

Cl

(107)

B i f o n a z o l e ( 1 0 9 ) is c l a i m e d t o b e r e m a r k a b l y n o n - t o x i c a n d i s m a r k e t e d a s a t o p i c a l a n t i f u n g a l a g e n t o v e r s e a s . It c a n b e c o n v e n i e n t l y s y n t h e s i z e d i n t h e b y n o w f a m i l i a r w a y b y r e d u c t i o n of r > p h e n y l b e n z o p h e n o n e ( 1 0 8 ) with b o r o h y d r i d e , c o n v e r s i o n to the c h l o r i d e w i t h thionyl chloride, and then imidazole displacement to bifonazole (109) [39].

(108)

(109)

E n o x i m o n e (113), an i m i d a z o l i n o n e - c o n t a i n i n g m o l e c u l e , is a cardiotonic m o l e c u l e like i m a z o d a n a b o v e . Its m e c h a n i s m o f a c t i o n i s still n o t e s t a b l i s h e d b u t it is r e p o r t e d t o b e a p o t e n t

94


inhibitor of a cAMP phosphodiesterase isoenzyme It is synthesized by Fnedel Crafts acylation of 4 methyhmidazolm 2 one (111) with 4 fluorobenzoyl chlonde (110) to give unsymmetrical bi aryl ketone 112 The synthesis is concluded by a nucleophihc aromatic displacement reaction with methyl mercaptan to give enoximone (113) [40]

on

s^JK

^M.

A^

"Y" ~ (110)

(111)

012)

A related agent piroximone (116) is also an active cardiotonic agent by virtue of marked strengthening of the toue oi the heart beat and reducing after load Its synthesis is accomplished b\ I ncdel Crafts acylation oi 4 tthyhmida/ohn 2 one (115) with the acid chlonde oi lsonicouiue dud (114) [411

+

(114)

.El HN NH

_

(115)

o J[ 1A N^J HN NH (U6)

Amflutia/ole (119) is an isothiazole containing gout suppressant by virtue of its inhibitory action against xanthine oxidase, an important enzyme in the catabohsm of uric acid Gout is charactenzed by crystallization of sharp needles of excess unc acid in joints Decreasing its metabolic formation relieves this condition The drug is synthesized by reaction of the tosvloxime denvative 117 with methyl thioacetate to give 118 by addition elimination The latter undergoes intramolecular cychzation between the active methylene and the electrophihc CN moiety on treatment with base to produce amflutiazole [42]

(117)

(118)

Five Membered Ring Heterocycles

95

Faneti/ole (122) is a biological response modifier with significant immunosuppressant activity It can be synthesized by conversion of 2 phen\lethylamine (120) with ammonium thio cyanate to the corresponding thiourea analogue 121 The synthesis of faneti/ole (122) concludes by thiazole ring formation of 121 by reaction with phenacylbromide Thus its synthesis involves use of the classic Hantzsch procedure in which a bromoacetone analogue and an appropnate thio urea derivative are reacted [43]

(120)

[12\)

(122)

Another thnzole containing diug ni/atidme (128) is in intagomstot histamine at Ho nxcptois and thus is in intiulctr aûnt rtl Uui to unutidmt It h is shown oril activity in the clinic Ni/atidme can be synthesized by re iction of ethyl bromopyruvate (124) with dimeth \laminothioacetamide (123) Presumably h ihde displacement is followed by cyclodehydration in producing thiazole 125 Hydride ieduction uid HBr mediated displacement with 2 aminoethan ethiol gives 126 This last is rtacted with 1 (meth\lthio) 2 nitro 1 N methylethylene (127) to give ni/atidme (128) [44] yNH2 Me NCI I—^

+

BrCHCOCO2Et

NL^COEt Mc2NCH2—Ues

3 THREE HETEROATOMS fnazolone containing nefa/odone (148) is an antidepressant agent It is of particular interest because of its unusual spectrum of receptor interactions It interacts with both adrenergic and serotonin receptors but not with muscannic acetylchohne receptors or monoamine oxidase Of the various syntheses one of the more interesting starts with 2 ethyloxazolme (143) which reacts w ill phenol to pioduce the propionamidt ether 144 1 his is converted to tht lminochlonde (145) with phosgene Reaction of the latter with methyl carbazate produces 146 The latter cychzes to the triazolone 147 by ester amide mteichange in base 1 he sequence terminates with alk> lation of 148 by 1 (3 chlorophenyl) 4 0 chloropiopyl)piperazine [511 RLFfRLNCLS 1 M N Cayen, R Gonzalez, E S Fernandini E Giesehn, D R Hicks, M Krarm, and D Dvormk, Xenobiotica, 16, 2il (1986) 2 C R Rasmussen, U S Patent 4 211,867 (1980) via Chem Abstr , 93 204 452d (1980) 3 I J Ward, Patentschnft (Switz ), 641,160 (1984) via Chem Abstr , 101 23,328f (1984) 4 F Franco, R Greenhouse, and J M Muchowski, / Org Chem , 47, 1682 (1982) 5 A A Patchett,E Hams E W Tristram, M J Wyvratt, M T Wu, D Taub,E R Peter son, T J Ikeler, J ten Broeke L G Payne, D L Ondeyka, E D Thorsett, W J Greenlee, N S Lohr, R D Hoffsommer, II Joshua, W V Ruyle, J W Rothrock, S D Aster, A L Maycock, F M Robinson, R Hirschmann, C S Sweet, E H Ulm, D M Gross T C Vassil,andC A Stone, Nature 288,280(1980) 6 M J Wyvratt, E W Tristram, T J Ikeler, N S Lohr, H Joshua, J P Spnnger, B H Arison, and A A Patchett,/ Org Chem ,49, 2816 (1984) 7 A A Patchett.E E Hams, M J Wyvratt, and E W Tristram, Eur Pat Appl, 12,401 (1980) via Chem Abstr , 95 25,634j (1981) 8 A Miyake,K Itoh,andY Oka, Chem Pharm Bull, 34,2852 (1986) 9 M T Wu, A W Douglas, D L Ondeyka, L G Payne, T J Ikeler, H Joshua, and A A Patchett, J Pharm Sci, 74, 352 (1985) 10 M A OndettiandJ Krapcho, U S Patent, 4,316,906 (1982) via Chem Abstr 96 218,225f (1982)


99

11. E. H. Gold, B. R. Neustadt, and E. M. Smith, U. S. Patent, 4,470,972 (1984) via Chem, Abstr., 102:96,083c (1984). 12. V. Teetz, R. Geiger, R. Henning, and H. Urbach, Arzneim.-Forsch., 34, 1399 (1984). 13. V. Teetz, R. Geiger, and H. Gaul, Tetrahedron Lett., 25, 4479 (1984). 14. S. Mzengeza, C. M. Yang, and R. A. Whitney, /. Am. Chem. Soc, 109, 276 (1987). 15. R. B. Silverman and M. W. Holladay, J. Am. Chem. Soc, 103, 7357 (1981). 16. J. E. Baldwin, L. I. Kruse, and J.-K. Cha, J. Am. Chem. Soc, 103, 942 (1981). 17. J. E. Baldwin, J. K. Cha, and L. I. Kruse, Tetrahedron, 41, 5241 (1985). 18. G. D. Diana, R. C. Oglesby, V. Akullian, P. M. Carabateas, D. Cutcliffe, J. P. Mallamo, M. J. Otto, M. A. McKinley, E. G. Maliski, and S. J. Michalec, J. Med. Chem., 30, 383 (1987).

6

Six-Membered Heterocycles

As noted earlier, benzeneringsin biologically active compounds act largely as centers of defined electron density and as rigid nuclei for attachment of groups which have pharmacophoric action. Aromatic, monocyclic heterocycles often play a similar role. It is, for example, often possible to replace a benzene ring by a heterocycle such as pyridine without markedly affecting biological activity even in face of the presence of the unshared basic pair of electrons in pyridine. There are, however, cases where the heterocycle forms an indispensable part of the chromophore; for example, a dihydropyridine is, as far as is known, an absolute requirement for calcium channel blocking activity. In a similar vein, 2~amino-4-pyridones form an essential part of those histamine H-2 antagonists lacking the cyano- or nitroguanidine functions. L PYRIDINES A rather simple pyridine derivative shows activity as an immunoregulator. Alkylation of 4-chloromethylpyridine (2), available from 4-picoline (1), with l-hydroxyethane-2-thiol affords ristianol (3) [1]. One of the first classes of compounds which provided practical chemical control of hypertension was guanidine derivatives. The mechanism of action of these agents, peripheral sympathetic blockade, resulted in side effects which led the class to be superseded by agents which acted by more acceptable means. A recent guanidine antihypertensive, which incorporates a cyano group on one of the nitrogens - making the functionality more akin to a thiourea - may act by a different mechanism. Condensation of the isothiocyanate 4 from 4-aminopyridine with the appropriate sec-amine gives thiourea 5. Treatment of that intermediate with a mixture of triphenylphos101

102


phine, carbon tetrachloride, and triethylamine leads to the unsymmetrical carbodiimide 6. Addition of cyanamid affords pinacidil (7) [2]. CH2SCH2CH2OH (D;X = H (2); X = Cl S Me II I -NHCNHCHCMc3 (5)

(3) Me *~

N -

N

(6) Me -C-S

(4)

\y-NHCL (7)

Replacement of one of the benzeneringsin a fenamic acid by pyridine interestingly leads to a compound which exhibits antihypertensive rather than antiinflammatory activity. Preparation of this agent starts with nucleophilic aromatic substitution of anthranilic acid (8) on 4-chloropyridine. The product (9) is converted to its acid chloride (10), and this is condensed with piperidine. There is thus obtained ofornine (11) [31. The majority of analgesics can be classified as either central or peripheral on the basis of their mode of action. Structural characteristics usually follow the same divisions; the former show some relation to the opioids while the latter can be recognized as NSAID's. The triamino pyridine 17 is an analgesic which does not seem to belong structurally to either class. Reaction of substituted pyridine 13 (obtainable from 12 by nitration ) with benzylamine 14 leads to the product from replacement of the methoxyl group (15). The reaction probably proceeds by the addition elimination sequence characteristic of heterocyclic nucleophilic displacements. Reduction of the nitro group with Raney nickel gives triamine 16. Acylation of the product with ethyl chloroformate produces flupirtine (17) [4].


103

v^/COR NH2 (8)

(9) ; R = OH (10); R = C1

+ F—/^—CH2NH2 (n);X = H2 (13);X = NO2

*~ F—C~\—CH2NH—i^—NR2 ( i5,. R

(14)

(11)

=o (16);R = H

N 2

"

d—C~\~ NHCO2Et NH2 (17) The causative agents of malaria, the plasmodia, have, in common with virtually all other microorganisms, developed tolerance to the chemotherapeutic agents which at one time seemed ready to wipe them from the face of the earth. Successful treatment of malaria has thus demanded a continuing effort to develop ever newer agents. The incentive for developing new antibiotics lies in the enormous market for these drugs in the so-called developed nations. No such incentive exists for research on new antimalarial drugs, since this is largely a disease of the third world. This accounts in part for the few entries in this category in recent volumes in this series. One of the newer drugs for treating malaria, enpiroline (22), interestingly incorporates a great many structural features of quinine (23). It is of note that this drug owes its existence to research sponsored by the U.S. Army's Walter Reed Institute rather than a pharmaceutical company. The starting nicotinic acid derivative 19 is the product of the rather complex condensation of keto-acid 18 with 1-trifluoroacylmethylpyridinium bromide and ammonium acetate patterned on the work of Kroehnke [5]. Condensation of the product with an excess of 2-lithiopyridine gives the ketone 20.


104

Treatment with sodium borohydride leads to the alcohol 21. Catalytic reduction of the product in the presence of acid leads to preferential hydrogenation of the monosubstituted ring. Selectivity is probably due to the fact that pyridine rings must be protonated in order to be reduced; the larger number of electron withdrawing substituents on the disubstituted ring will reduce basicity of the nitrogen and favor protonation of the alternate pyridine nitrogen. Separation of the R*,R* isomer from the diastereomeric mixture affords enpiroline (22) [6].

Br" NH4OAc

fcCH=CHCO2H (18)

CO2H

F,C

F3O

(20);R = O (21); R « H, OH

(19)

McQF,O

(23)

(22);

R*,R*

T h e search for nonsedating H-1 antihistamines met its first success in terfenadine (see 198, Chapter 2). A different approach aimed at keeping such agents out of the C N S , by prevent-


105

ing their crossing the blood brain barrier, consists in converting some known antihistamine to a zwitterion by incorporating a carboxyl group. Application of this strategy to triprolidine (28 minus the pendant acrylic acid chain) results in the antihistamine acrivastine (28). Synthesis of this compound starts by reaction of the mono lithio derivative from pyridine 24 with r>toluonitrile. Hydrolysis of the intermediate imine affords the benzophenone 25. Condensation of the carbonyl group with the ylide from triphenyl(2-N-pyrrolidinoethyl)phosphonium chloride gives the intermediate 26, probably as a mixture of geometric isomers. The remaining halogen on the pyridine ring is then converted to the lithio reagent by halogen metal interchange with an alkyl lithium. Condensation of the resulting organometallic with dimethylformamide gives, after hydrolysis, the aldehyde 27. This is again subjected to a Wittig type reaction, this time with the ylide from triethylphosphono acetate. This last reaction leads almost exclusively to the trans double bond isomer. Saponification of the ester group affords the amino acid. Separation of the E,E isomer affords acrivastine (28) [7].

.Me

• B xX yX Xr J

(26)

(27)

(28)

106

Six-Membered Heterocycles 2. DfflYDROPYRIDINES

The calcium channel blocking agents, typified by the dihydropyridine nifedipine, were approved initially for use in cases of atypical, nonexercise induced angina. Continued clinical investigations have shown these agents to be useful for a variety of additional indications, including all types of angina as well as hypertension. This resounding commercial success has spurred work in numerous laboratories in order to develop their own proprietary dihydropyridine calcium channel blocking agents. The Hantsch pyridine synthesis provides the final step in the preparation of all dihydropyridines. This reaction consists in essence in the condensation of an aromatic aldehyde with an excess of an acetoacetate ester and ammonia. The need to produce unsymmetrically substituted dihydropyridines led to the development of modifications on the synthesis. (The chirality in unsymmetrical compounds leads to marked enhancement in potency.) Methyl acetoacetate forms an aldol product (30) with aldehyde 29; conjugate addition of ethyl acetoacetate would complete assembly of the carbon skeleton. Ammonia would provide the heterocyclic atom. Thus, application of this modified reaction affords the mixed diester felodipine 31 [8].

O MeCCH2CO2Et NHa

a

g^Me (31)

(32)


107

CHO (33)

(34)

(35); R1 = CHMe2 ; R2 = Me (36);R1 = Et;R2 = Et

(37) R a n d o m incorporation of two different acetoacetates can also be avoided by converting one of the acetoacetates to a derivative which carries the future pyridine nitrogen. F o r example, treatment of ethyl acetoacetate with a m m o n i a gives the corresponding p-aminocrotonate 32. T h e aldehyde (34) required for preparation of such an unsymmetrical c o m p o u n d is prepared by reaction of the product from direct metallation of 3 3 with dimethylformamide. Condensation of that aldehyde with methyl acetoacetate and the p-aminocrotonate from isopropyl acetoacetate leads to i s r a d i p i n e (35) [9]. T h e same aldehyde with ethyl acetoacetate and the p-aminocrotonate from ethyl acetoacetate gives d a r o d i p i n e (36) [10]. In much the same vein, condensation of the benzaldehyde 37 with methyl acetoacetate and its P-aminocrotonate derivative affords r i o d i p i n e (38) [in. Activity is apparently retained w h e n the ring nitrogen is alkylated as in f l o r d i p i n e (42). Aldol condensation of the benzaldehyde 3 9 with ethyl acetoacetate gives the unsaturated ester 40. T h e nitrogen containing reaction partner 4 1 is obtained by condensation of 32 with 2- morpholinoethylamine. Reaction of 40 with 4 1 leads to flordipine (42) [12]. The methyl groups adjacent to the pyridine nitrogens can also be modified without changing calcium channel blocking activity. T h e most significant change involves replacement of methyl by a nitrile group. Hantsch type condensation of the nitrobenzaldehyde 4 3 with methyl acetoacetate and the vinyl amine 44 from isopropyl 3-cyano-3-ketopropionate leads directly to nilvadipine(45)[13].


108

EtO2CN Jl

F3C

(32)

HNCH2CH2N/

\)

(41)

CH2CH2N

O

(42) T h e preparation of an analogue containing an oxygenated methyl group starts by the displacement of halogen on the chloro acetoacetate 47 by the protected ethanolamine derivative 46. Condensation of the product (48) with the vinyl amine derivative of methyl acetoacetate and o-chlorobenzaldehyde gives the dihydropyridine 49. Removal of the benzyl protecting groups by catalytic hydrogenation, affords amlodipine (50) [14]. T h e search for opioid analgesics which show reduced addiction liability has centered largely on benzomorphan and morphinan derivatives. S o m e research has, however, been devoted to derivatives of the structurally simpler meperidine series. T h e preparation of one such compound, picenadol (59), starts with the reaction of N-methyl-4-piperidone with the lithium derivative from m-methoxybromobenzene. Dehydration of the first formed carbinol 5 1 gives the intermediate 5 2 . Deprotonation by means of butyl lithium gives an anion which can be depicted in the ambident form 5 3 . In the event, treatment of the anion with propyl bromide gives the product 54 from reaction of the benzylic anion. Treatment of that product, which now contains an eneamine function,


cL CO Me M 2 2

109

+

NCC=CHCO2CHMe2

O (C6H5CH2)2NCH2CH2OH

ClCH2CCH2CO2Et (47)

(46)

O (C6H5CH2)2NCH2CH2OCH2CCH2CO2Et CH2OCH2CH2NR2 (48)

(49); R = CH2C6H5 (50); R = H

under Mannich reaction conditions (formaldehyde, dimethylamine) leads to the aminomethyl derivative 56. It m a y be speculated that the reaction proceeds initially through the methyl carbinol 5 5 . Hydrogenolysis of 56 involves initially removal of the allylic amino group to afford a transient intermediate such as 5 7 , followed by reduction of the enamine to produce 5 8 . Demethylation of the phenolic ether by m e a n s of hydrogen bromide and fractional crystallization to obtain the derivative with cis configuration of the two ring alkyl groups completes the synthesis of the analgesic picenadol (59) [15]. Reaction of dimethylformamide with dimethyl sulfate leads to the highly reactive O methyl ether 60. Exposure of this reagent to n-octylamine leads to the amidine 6 1 . A n exchange reaction between this last intermediate and the piperidine derivative 62, results in displacement of dimethylamine by the piperidine nitrogen. There is thus obtained the gastric antisecretory agent fenoctimine (63) [16]. T h e structural fragment represented by 62 is often used in H - l antihistamines; the fact that those c o m p o u n d s often exhibit some anticholinergic activity m a y account for the activity of 6 3 .


110 OMe

OMe NMe

NMe (51)

(52)

OMe

OMe NMe

(53)

OMe

OMe

NMe

(57)

(60)

(63)

(58); R = Me (59); R = H


111

A rather more complex compound, levocabastine (72), is described as an extremely potent, selective, H-l antihistaminic agent. The presence of a free carboxyl group suggests that this compound may have difficulty in crossing the blood brain barrier and may thus show reduced sedating activity. One leg of the convergent synthesis starts with the double conjugate addition of ethyl acrylate to the anion from r>fluorophenylacetonitrile (64). Base catalyzed cyclization of the product (65) affords the keto ester 66. Decarboethoxylation of that intermediate gives the cyanocyclohexanone 67. Esterification of the carboxyl group in the optically active meperidine-related compound 68 with benzyl chloride leads to ester 69. The tosyl protecting group is then removed by means of electrolytic reduction. Condensation of the ketone 67 with secondary amine 70, under reductive alkylation conditions affords predominantly (9:1 ratio) the intermediate 71, which is purified by recrystallization. The trans stereochemistry of the cyclohexane amino group and the aromatic ring result from the tendency to form equatorial nitrogen in the reduction of the intermediate imine. The benzyl protecting group is then removed by means of a second, more drastic hydrogenation step. The product of that reaction is then the levorotatory isomer, levocabastine (72) [17].

-CH2CN

\:H2CH2CO2Et

(64)

(66); R = CO2Et (67); R = H

(65)

TsN (68); (3£, 4£)

(69); R = Ts (70); R = H

,Me

NO

(72)

(71)


112

Treatment of the piperidine 74, obtainable from an aminonitrile such as 73, under Nmethylation conditions leads to the dimethylamino derivative 75. The carbobenzoxy protecting group is then removed by catalytic hydrogenation. Reaction of the resulting secondary amine 76 with cyclohexene oxide leads to the alkylated trans aminoalcohol. There is thus obtained the antiarrhythmic agent transcainide (77) [18].

CN Me

RN

Me

(73)

(74); R = H (75); R = Me

CONH Q - C X ; •NMC2

(77)

(76)

3. PYRIMTDINES T h e discovery of the histamine H-2 blocking agents has virtually revolutionized the treatment of ulcers of the upper GI tract. Drugs such as cimetidine (78) and ranitidine (89) have proven so safe and effective that serious consideration is being given to granting approval for sale of these without prescriptions. This very success has engendered considerable work aimed at exploring the limits of the S A R . It was found early on that the cyanoguanidine function could be replaced by a pyrimidone (see The Organic Chemistry

of Drug Synthesis,

V o l u m e 3). M o s t of the recent work

in this series h a s apparently focused on such pyrimidones. Catalytic reduction of the nitrile 7 9 in the presence of semicarbazide affords initially the semicarbazone of 80. Hydrolysis-interchange, for example in the presence of pyruvic acid, gives the aldehyde 80. Condensation with the half ester of malonic acid leads to the acrylic ester 8 1 ; the double bond is then removed by means of catalytic reduction (82). B a s e catalyzed reaction of the


113

ester with ethyl formate gives the corresponding formyl derivative 83. Condensation of this beta dicarbonyl compound with N-nitroguanidine leads to the pyrimidone 84. The reaction involves, as expected, the more nucleophilic (i.e., nonnitrated) guanidine nitrogens; the product, in addition, contains a built in good leaving group. Thus, displacement of the leaving group with pyridylpropylamine 85, affords the H-2 blocker icotidine (87) [19]. The same reaction using the bromopyridine 86 as the nucleophile, gives the H-2 blocker temelastine (88) [20].

Me.

NCN C —NHMe

.CH (78)

CH=CHCO2R NT

(79)

, X X

(80) I

:HOH

Me (83)

I

(82)

O N CH2CH2CH2CH2NH2

•Me

¥ ^^N H (84)

( 8 5 ) ; R ! = H ; R 2 = OMe (86); R1 = Br; R2 = Me

H (87)

H (88)


114

Ranitidine (89), in which the imidazole ring has been replaced by a furylmethyl moiety has proven a particularly successful drug. Several pyrimidones thus incorporate that ring system. Repetition of the scheme used to prepare the pyrimidone intermediate 84, starting with the oxygenated pyridine 90 instead of 79, affords the key synthon 91. Displacement of the leaving group by the primary amine from the furyl alkylamine 92 leads to 93. Treatment of this last compound with dry hydrogen chloride leads to cleavage of the pyridine O-methyl ether and formation of the corresponding hydroxy derivative. This function is more usually depicted as a pyridone. There is thus obtained donetidine (94) [21].

(89)

.CN

(90)

MeHN^

OMe

(92)

(91)

CH2SCH2CH2NH ^

N H

(93)

CB MejNCHf

(94)


115

An alternate scheme for preparing these compounds starts with a prefabricated pyrimidone ring. Aldol condensation of that compound (95), which contains an eneamide function, with pyridine-3-aldehyde (80), gives the product 96. Catalytic hydrogenation gives the product of 1,4 reduction. The resulting pyrimidinedione, of course exists in the usual tautomeric keto (97a) and enol (97b) forms. Reaction with phosphorus oxyxchloride leads to the chloro derivative 98. Displacement with methoxide gives 99. Reaction of this last intermediate with the furylalkylamine derivative 92 leads to the H-2 blocker lupitidine (100) [22]. O o

x_v

H

-Me

HN"" (95)

(96)

N

H

(97a)

^

r

-

Q

-

™

(97b);R = OH (98); R = Cl (99); R - OMc o

H (100) Pyridones such as amrinone (101) and milrinone (102) have proven to be very effective cardiotonic agents. It is of interest that activity is retained when an additional nitrogen is inserted into the ring to form a pyrimidone. Condensation of amidine 103 with intermediate 104 can be visualized as involving initially addition elimination of basic nitrogen to the highly electrophilic double bond and loss of the good, though odiferous, leaving group, methyl mercaptide (105). Cyclization by replacement of the ester methoxide group will then give pyrimidone 106. Re-


116

placement of the remaining thiomethyl group by 3-methylaminopyridine affords the cardiotonic agent pelrinone (107) [23],

(102)

(101) HN

MeSw i2 MeS (103) (104)

SMe H (105)

O

NV 'N-^c H (107)

CN H (106)

Chemotherapy of diseases caused by microorganisms is at least conceptually very straightforward since it depends on deep seated metabolic differences between eukariotic and prokariotic species; ithas as consequence usually been possible to identify compounds which are uniquely toxic to disease causing organisms. In contrast to this, viral and neoplastic diseases have in common a derangement of the host cells' mechanism for replication; this process is taken over by the virus in one case and is no longer under tight genetic control in the other. The clean differences in metabolism simply do not exist. The fact that both disease types involve nuclear processes and thus also nucleosides has led to a long-term research effort to synthesize modified purines and pyrimidines in the hope that some subtle differences may exist between healthy and diseased cells. The success of the modified nucleoside, acyclovir, suggests that this approach has considerable merit.


117

One such compound, bropirimine (112), is described as an agent which has both antineoplastic and antiviral activity. The first step in the preparation involves formation of the dianion 108 from the half ester of malonic acid by treatment with butyllithium. Acylation of the anion with benzoyl chloride proceeds at the more nucleophilic carbon anion to give 109. This tricarbonyl compound decarboxylates on acidification to give the beta ketoester 110. Condensation with guanidine leads to the pyrimidone 111. Bromination with N-bromosuccinimide gives bropirimine (112) [24]. O CO2H II / -C—CH N CO2Et

,CO2H CH,

COoEt

(109)

O

I -c—cH2co2a (112)

(111)

(110)

Compounds prepared from naturally occurring nucleosides are of course more closely related to genetic material and may have a better chance of interacting with infected cells. Mercuration of the 2'~deoxyuridine 113 leads to the organometallic derivative 114; reaction of that with ethylene in the presence dilithio palladium tetrachloride gives the alkylation product 115; this is reduced catalytically in situ. There is thus obtained the antiviral agent edoxudine (116) [25]. Todays' most famous virus is probably that which is responsible for AIDS. Though an actual cure is not yet in sight, a drug which slows progression of the disease by inhibiting replication of the virus was recently introduced. The published synthesis for the drug, known trivially as AZT, startsfromthymidine, possibly accounting for its very high cost when initially made available. Thus, treatment of thymidine (117) with chloropentafluorotriethylamine leads to the product from displacement of the sugarringhydroxyl by the hydroxyl from the pyrimidone eno-

118


late (118). The potent nucleophile, azide, serves to open the somewhat strained bridgingring;this involves a second inversion and restoration to the natural stereochemistry. There is thus obtained zidovudine (119), formerly known as AZT [26]. O

(118)

O

(119)

4. PIPERAZINES Yet another nonsedating zwitterionic H-1 antihistamine consists of the product from metabolism of the terminal hydroxyl of the potent antihistamine hydroxyzine terminating in hydroxymethyl instead of a carboxylic acid. This compound, cetirzine (123), can be obtained in straightforward fashion by alkylation of the monosubstituted piperazine 120 with halide 121, via the amide 122 [27]. A number of diarylmethyl alkylpiperazines, such as, for example lidoflazine, have found use as coronary vasodilators for the treatment of angina. The most recent of these interestingly incorporates a 2,6-dichloroaniline moiety reminiscent of antiarrhythmic agents. Treatment of the piperazine carboxamide 124 with acetone leads to formation of the nitrogen analogue of an acetal, the aminal 125. Alkylation of the remaining secondary nitrogen with chloroamide 126 leads to the intermediate 127. Exposure to aqueous acid leads to hydrolysis of the aminal function


119

and restoration of the secondary amine 128. Alkylation of that center with iodide 129 followed by N-demethylation leads to the formation of mioflazine (130) [28]. cu NH + C1CH2CH2OCH2CONH2 (120)

HN NH MeHN r o (124)

NCH2CH2OCH2COR

(122); R = NH2 (123); R = OH

(121)

? C1CH C H 2N (126)C1

\ NH

7

0 (125)

H : CHC C I2 2H 2H

(129) (127)

oa N CHC H2N a H2 (130) The anxiolytic agent buspirone (131) is notable for the fact that it does not interact with the receptor for the benzodiazepines. This difference in biochemical pharmacology is reflected in the fact that buspirone (131) seems to be devoid of some of the characteristic benzodiazepine side effects. The spiran function is apparently not required for anxiolytic activity. Alkylation of 3,3-dimethylglutarimide with dichlorobutane in the presence of strong base yields the intermedi-

120


ate 132; treatment of that with ethylenediamine leads to 133. Use of this diamine to alkylate 2chloropyrimidine proceeds at the terminal amino group to afford the open chain compound 134. Bisalkylation with 1,2-dichloroethane leads to ring closure to a piperazine. There is thus obtained gepirone (135) [29]. / NH X

—

Me/N

/ NR

—

t e

/ .—^ 1 N(CH2)4NHCH2CH2NH —(' x

0 (132);R = (CH2)4C1 (133); R « (CH2)4NHCH2CH2NH2

o / (134) /

(135)

(131) 5. MISCELLANEOUS COMPOUNDS A piridazine ring forms the nucleus for a rather unusual nontricyclic antidepressant Condensation of the keto ester 136 with hydrazine leads to the cyclic hydrazide 137. Oxidation, for example with bromine, gives the corresponding pyridazone 138. The oxygen is then replaced by chlorine by reaction with phosphorus oxychloride. Displacement of the halogen in 139 with N-ethylaminomorpholine affords minaprine 140 [30], Reaction of 2,3-dichlorobenzoyl chloride with cyanide ion leads to the corresponding benzoyl cyanide (141). Condensation of that reactive intermediate with aminoguanidine 142 leads to the hydrazone-like product 143. Treatment with base results in addition of one of the guanidine amino groups to the nitrile function and formation of the 1,2,4-triazine ring. The product, lamotrigine (144), is described as an anticonvulsant agent [31].


121

(138)

(137)

(136)

(139)

(140)

•O

(141)

+

(142)

(143)

(144)

Replacement of heterocyclic rings in nucleosides by ring systems which do not occur in nature represents another approach to c o m p o u n d s which may have activity against viral and neoplastic diseases. One of the early successes in this category involves replacement of a pyrimidine ring by a triazine. T h e synthesis starts with a now classical glycosidation of a heterocycle as its silylated derivative (146) with a protected halosugar (145), in this case a derivative of arabinose


122

[32]. Hydrogenation of the product 147 removes the benzyl protecting groups and at the same time reduces the triazine to its dihydro derivative 148. A roundabout scheme is required for dehydrogenation due to the sensitivity of the intermediates. The product is thus converted to its silyl ether 149; exposure to air results in oxidation and desilylation. There is thus obtained the antineoplastic agent fazarabine (150), also known as ara*A C.

NHSiMe.,

BzO

(146)

BzO

(147)

OH tin (150)

(149)

(148)

Alkylating agents represent the oldest class of antineoplastic agents. These compounds, whose actions are somewhat indiscriminate, disrupt genetic material by forming covalent bonds with the bases in DNA. Polydentate alkylating agents may be more effective, it is thought, due to their ability to crosslink adjacent DNA chains. One such compound is available from the reaction of cyanuric acid (151) with epichlorohydrin [33]. Each of the three epoxypropyl side chains contains a chiral center. The product will thus consist of a mixture of 4 enantiomers (2 diastereomers) [34]. The drug, teroxirone (152), in fact consists of the separated racemic (R S^ R S^ S R)-isomer.


123

(151)

-b» (152); &*,£*, S.* REFERENCES

1. J. G. Lombardino and C. A. Harbert, Belg. Pat., 890,374 (1982) via Chem, Abstr., 97: 23,635c (1982). 2. H. J. Petersen, C. K. Nielsen, and E. A. Arrigoni-Martelli, /. Med. Chem., 21, 773 (1978). 3. D. M. Bailey, Eur, Pat. AppL, 98,951 (1984) via Chem. Abstr.1102: 203737t (1985). 4. W. Orth, J. Engel, P. Emig, G. Scheffler, and H. Pohle, Ger. Offen., 3,608,762 (1986) via Chem. Abstr., 106: 50,057b (1987). 5. F. Kroehnke, W. Zecher, J. Curtze, D. Drechsler, K. Pfleghar, K.-E. Schnalke, and W. Weis, Angew. Chem., 14, 811 (1962). 6. A. B. Ash, M. P. LaMontagne, and A. Markovac, U.S. Patent, 3,886,167 (1975) via Chem. Abstr., 83: 97,041p (1975). 7. G. G. Coker and J. W. A. Findlay, Eur. Pat. AppL, 85,959 (1983) via Chem. Abstr., 100: 6,345w (1984). 8. P. B. Berntsson, A. I. Carlsson, J. O. Gaarder, and B. R. Ljung, Swed. Patent, 442,298 (1985) via Chem. Abstr., 106: 4,878x (1987). 9. P. Neumann, Eur. Pat. AppL, 0,150 (1979) via Chem. Abstr., 92: 94,404j (1980). 10. M. Heitzmann, Patentschrift (Switz.), 661,728 (1987) via Chem. Abstr., 107: 236,717t (1987). 11. V. V. Kastron, G. J. Dubur, V. D. Shatz, and L. M. Yagupolsky, Arzneim- Forsch., 35, 668 (1985). 12. F. C. Huang, C. J. Lin, and H. Jones, Eur. Pat. AppL, 109,049 (1984) via Chem. Abstr., 101:110,747k (1984). 13. Y. Sato, Ger. Offen., 2,940,833 (1980) via Chem. Abstr., 93, 220,594g (1980). 14. S. F. Campbell, P. E. Cross, and J. K. Stubbs, US. Patent, 4,572,909 (1986) via Chem. Abstr., 105: 42,661h (1986).

124


16. G. Feth and J. E. Mills, U.S. Patent, 4,499,274 (1985) via Chem. Abstr., 102: 166,624f (1985). 17. R. A. Stokbroekx, M. G. M. Luyckx, J. J. M. Willems, M Janssen, J. O. M. M. Bracke, R. L. P. Joosen, and J. P. Van Wauwe, Drug Dev. Res., 8, 87 (1986). G. Vanden Bussche, Drugs of the Future, 11, 841 (1986). 18. G. H. P. Van Daele, M. F. L. De Bruyn, and M. G. C. Verdonck, Eur. Pat. AppL, 121,972 (1984) via Chem. Abstr., 102: 149,116z (1985). 19. T. H. Brown, G. J. Durant, and C. R. Ganellin, Can., 1,106,375 (1981) via Chem. Abstr., 96: 35,288j (1982). 20. G. S. Sach, Eur. Pat. AppL, 68,833 (1983) via Chem. Abstr., 99: 22,483f (1983). 21. B. M. Adger and N. I Lewis, Eur. Pat. AppL, 141,560 (1985) via Chem. Abstr., 103: 178,273z(1985). 22. B. L. Lam and L. N. Pridgen, Eur. Pat. Appl, 58,055 (1982) via Chem. Abstr., 98: 16,719a (1983). 23. J. F. Bagli, Eur. Pat. AppL, 130,735 (1985) via Chem, Abstr., 103: 6,358q (1985). 24. H. I. Skulnick, S. D. Weed, E. E. Eidson, H. E. Renis, W. Wierenga, and D. A. Stringfellow,7. Med. Chem., 28, 1864 (1985). 25. D. E. Bergstrom and J. L. Ruth, J. Am. Chem. Soc, 98, 1587 (1976). 26. M. Imazawa and F, Eckstein, /. Org, Chem,, 43, 3044 (1978). 27. E. Baltes, J. De Lannoy, and L. Rodriguez, Eur. Pat. AppL, 58,146 (1982) via Chem. Abstr., 98: 34,599r (1982). 28. G. Van Daele, Eur. Pat. AppL, 68,544 (1983) via Chem. Abstr., 99: 22,493j (1983). 29. V. Aguirre Ormaza, Span., 550,086 (1986) via Chem. Abstr., 107: 39,628p (1987). 30. C. Wermuth, H. Davi, and K. Biziere, Eur. Pat. AppL, 72,299 (1983) via Chem. Abstr., 99: 105,265n(1983). 31. M. G. Baxter, A. R. Elphick, A. A. Miller, and D. A. Sawyer, Can., 1,133,938 (1982) via Chem. Abstr., 98: 89,397d (1983). 32. M. W. Winkley and R. K. Robins, J. Org. Chem., 35,491 (1970). 33. D. Joel and H. Becker, Plaste Kautsch., 23, 237 (1976); via Chem. Abstr., 85: 21,303w (1976). 34. M. Budnowski, Angew. Chem., 7, 827 (1968).

7

Five-Membered Ring

Benzofused Heterocycles

The vast majority of drugs contain an aromatic ring of some type or other as an important structural element. Biological systems are often quite forgiving in their acceptance of interchange of certain heteroatoms for carbon in such rings fe. g., N (benzene to pyridine), O (furan for pyrrole), S (thiophene for furan)] as shown by the popularity of the stratagem of bioisosteric replacement as an element of drug design. In this context, the heteroaromatic ring often appears to serve as a flat, electron rich framework upon which to attach functionality which may interact with specific receptors. These factors, the relative ease of construction of heterocycles, and the huge numbers of isomers this makes possible account for many of the very large family of drugs incorporating this structural feature. In such molecules the specific aromatic moiety is rarely essential for activity. As one consequence, the drugs discussed in this chapter display a very wide range of bioactivities. 1. BENZOFURANS Thromboxane A-2 has been implicated in a number of disorders of the circulatory system including coronary artery spasms, unstable angina pectoris, traumatic and endotoxic shock, and heart attacks. It is formed normally very near its receptors and is rapidly deactivated by metabolizing enzymes so circulating levels are quite low. Furthermore, it is opposed in its actions by the prostacyclins. When these controls are defective, pathology results and drugs can be the resort in attempts to restore the normal healthy balance. For one example, furegrelate (6) is a throm125

126

Five-Membered Ring Benzofused Heterocycles

boxane synthetase inhibitor which can reestablish homeostasis by banking down the synthesis of thromboxanes. A useful ancillary characteristic of this type of drug is that the precursor prostaglandin endoperoxides are largely diverted to the biosynthesis of prostanoids whose pharmacological actions antagonize those of thromboxanes. Thus, both of its major actions operate in the same direction and the drug has shown clinical value in some life-threatening conditions f 1]. One of the syntheses of furegrelate begins by catalytic hydrogenation (Pd/C) of 3-(4nitrobenzyl)pyridine (1) to the corresponding aminobenzylpyridine (2). This is followed by diazotization in the usual fashion; the diazoniurn salt is transformed to the corresponding phenol (3) by heating with hot aqueous acid. A formyl group is then introduced ortho to the phenolic function by use of hexamethylene tetramine in anhydrous trifluoroacetic acid (the Duff reaction). The key intermediate in this interesting transformation is believed to be the substituted benzylrnethylene imine formed by interception by the phenol of the highly reactive intermediate formed by partial depolymerization of the hexamethylene tetramine reagent. The resulting substituted benzylmethyleneimine is in turn thought to undergo double bond isomerization to the corresponding substituted benzaldehyde methylimine. Hydrolysis during workup gives the final product. Treatment with base and diethyl bromomalonate produces the desired benzofuran ring system of 5. It seems likely that this reaction proceeds by ether formation followed by cyclization to the 3-hydroxy-2,2-dicarbethoxydihydrofuran system which, on saponification undergoes decarboxylative ejection of hydroxyl giving aromatic ester 5. The synthesis concludes by saponification to furegrelate (6) [2].

(1)

(2)

(3)

(4)

(5); R = Et (6); R = H


12'

A somewhat related nonsteroidal antiinflammatory agent, furaprofen (14), is much more active as its S-enantiomer. Its patented preparation depends upon a chiral enzymic hydrolysis in a late step. One synthesis begins by ether formation between 2-bromophenol (7) and phenacyl bromide (8) to give the aryloxyacetophenone 9. Treatment of 9 with polyphosphoric acid leads to cyclodehydration to benzofuran 10. The later is converted to the Grignard reagent and condensed with methyl pyruvate to give tertiary carbinol 11. Deoxygenation is accomplished by sequential dehydration with tosic acid and subsequent hydrogenation to produce racemic ester 12. Saponification produces racemic furaprofen (13). The use of a hydrolytic enzyme from Bacillus subtilis to convert 12 to 13 produces the optically active drug [3J.

BrCH2CO—f~\

—*-

Vsssas/

B r

( 8 )

( 9 )

( 1 0 )

A m i o d a r o n e

d e p r e s s a n t

T h e

o r i g i n a l l y

i o d o n a t e d

o f

1 5

g o u s

u s e f u l

i s

s t e p s

( 1 6 )

i n

1 5

d e t a i l e d

[ 6 ] .

h a s

b e e n

t r e a t i n g

p a t e n t e d

p h e n o l

n o t

(

t h e

c e n t e r

v e n t r i c u l a r

p r o c e d u r e

w i t h

i n

t h e

o f

m u c h

i n t e r e s t

a r r h y t h m i a

c o n c l u d e s

a n d

b e c a u s e

m a n y

o f

b y

e t h e r i f i c a t i o n

2 - h a l o d i e t h y l a m i n o e t h a n e

t o

g i v e

b u t

t h e

s y n t h e s i s

o f

; R

( 1 4 ) ;

R

i t s

a n a l o g u e s

s i m p l y

r e f e r e n c e

)

( 1 3 ) ;

o f

a m i o d a r o n e

b e n z b r o m a r o n e

= =

H H ,

S - e n a n i i o m c r

a c t i v i t y

h a v e

a s

b e e n

a

c a r d i a c

p r e p a r e d

[ 4 ) .

b e n z o f u r a n - c o n t a i n i n g

( 1 6 )

[ 5 ] .

c o n t a i n s

T h e

s y n t h e s i s

c l o s e l y

a n a l o -

128


OH I *-o'

(15)

(16)

2. INDOLINES One of the many angiotensin-converting enzyme (ACE) inhibitors covered in this volume contains an indoline residue instead of a pyrrolidine. As such, it may be considered as a benzo analogue of the captopril series. The synthesis of pentopril (19) follows the classical amide forming condensation of hemiester 17 (itself prepared by alcoholysis of the corresponding 2,4-dimethylglutaric anhydride) with (S)-indoline-2-carboxylic acid (18), using l~[3~(dimethylamino)propyl)|~3~ethylcarbodiimide as condensing agent, in order to produce pentopril [7].

"**8

(17)

U - K V H

v>t

(18)

" I P * (19)

Saturation of the aromatic ring of pentopril analogues is also consistent with ACE inhibition as demonstrated by the oral activity of indolapril (23). The necessary heterocyclic component (21) can in principle be prepared by catalytic perhydrogenation (Rh/C, HOAc) of the corresponding indole. A single isomer predominates. The product is condensed by amide bond formation with the appropriate alanylhomophenylalanyl dipeptide ester 20 to give 22. Selective saponification to 23 could be accomplished by treatment with HC1 gas. Use of the appropriate stereoisomers (prepared by resolution processes) produces chiral indolapril [8].


(20)

129

(21)

(22);R = *-Bu (23); R = H

3. BENZOTHIOPHENES Tipentosin (28) contains a partially reduced benzothiophene ring system and is of interest as an antihypertensive agent. Its synthesis begins with epoxidation of 3-acetamidocyclopentene (24) followed by epoxide opening with phenol, and then deacetylation to give amino alcohol 25. The second half of the molecule is prepared from 1,3-cyclohexanedione (26) by sequential alkylation with chloroacetone, isopropylidenation of the product (to the bisenolether), and cyclization to a thiophene moiety with hydrogen sulfide. The latter reaction appears to involve a sequential double Michael addition of hydrogen disulfide with consequent elimination of the acetonide in the form of acetone and water. Synthesis of this component is concluded by aminomethylation via a Mannich reaction followed by reverse Michael loss of dimethylamine to give 27. The synthesis of (ipentosin (28) concludes by Michael addition of the amino group of 25 to the conjugated linkage of 27 [9].

OH (25)

(24)

•Me o (26)

.

+

(27)

27

—

r r P V C bHH

n

(28)

^ I T

130


4. BENZISOXAZOLES A diuretic of the phenoxyacetic acid class is brocrinat (35). Its synthesis begins with FriedelCrafts acylation of resorcinol dimethyl ether (29) with 2-fluorobenzoyl chloride to give unsymmetrical benzophenone 30. The ortho-phenolic ether moiety is cleaved selectively in this acylation reaction. The product is converted predominantly to its E-oxime analogue (31) in the usual fashion and then acetylated (Ac2O) to its acetyl ester 32. This synthesis of the benzisoxazole ring concludes by NaH treatment leading to cyclization to 33 involving an internal displacement reaction. Metallation with n-BuLi followed by bromination produced 34. Ether cleavage with pyridine hydrobromide followed by alkylation with NaH and ethyl bromoacetate and hydrolysis lead to the oxyacetic acid containing brocrinat (35). [10].

(29)

(30)

(31)

^ (33) Br

(34) (34)

(35) (35)

The anticonvulsant activity of some 1,3-benzisoxazoles was discovered in routine testing. One of the more interesting of the subsequent analogues prepared was zonisamide (39). One of its syntheses starts with l,2-benzisoxazole-3-acetic acid (36) which is brominated and subsequently decarboxylated to give 37. Displacement of halogen in 37 with sodium bisulfite interestingly

Five-Membered Ring B enzofused Heterocycles

131

proceeds by reaction on sulfur to produce l,2-benzisoxazole-3-methane sulfonic acid (38). Chlorination of 38 with phosphorous oxychloride to the corresponding sulfonyl chloride followed by reaction with ammonia gives the sulfonamide, zonisamide (39) [11]. Alternate syntheses are available [12].

CH2CO2H

(36)

CH2Br

(37)

CH2SO3Na a

CH2SO2NH2

2 3

(38)

(39)

5. BEN2OXAZOLES Eclazolast (41) is an antiallergy compound which inhibits release of mediators of allergy. One reported synthesis involves the simple ester exchange of methyl 2-benzoxazolecarboxylate (40) with 2-ethoxyethanol catalyzed by sulfuric acid [13].

(40)

(41)

6. BENZIMIDAZOLES The last step in one reported preparation of the antiviral agent enviradene (43) involves dehydration of 6-(l-hydroxy-l-phenylpropyl)-2-amino-l-isopropylsulfonyl-benzimidazole (42). The Eproduct predominates [14]. Precedent for this chemistry and a description of related intermediates can be found in Volume 3, p. 177 of this work.

132


SO2CHMe2

Me-^H

(42)

SC^CHMe,

(43)

Benzimidazoles bearing a carboxamide function at the 2-position have provided the nucleus for a significant n u m b e r of anthelmintic agents. (See Chapter 11 of V o l u m e 2 and Chapter 10 of V o l u m e 3 of The Organic

Chemistry

of Drug Synthesis

for examples.) T h e high rate at which

resistant strains of parasites have developed has led to the need for ever newer drugs. Preparation of dribendazole (46) begins by reaction of the acetate of 2,5-dinitroaniline with cyclohexylmethylthiol; the product from the unusual displacement of one of the nitro groups (45) is then reduced to the diamine. Reaction of this intermediate with N,N~dicarbomethoxy~§-methylthiourea leads to the cyclized product [15].

(44)

(45)

(46)

An analogous sequence leads to the anthelmintic agent, etibendazole (50). Reaction of the benzophenone 47, which can be obtained by acylation of o-nitroaniline with £-fluorobenzoyl chloride, with ethylene glycol leads to acetal 48. Sequential reduction of the nitro group and cyclization of the resulting diamine (49) with N,N-dicarbomethoxy~S~methylthiourea gives the benzimidazole etibendazole (50) f 16].

v o (47)

(48); X = NO2 (49); X = NH2

(50)


13 3

The discovery of the antiulcer activity of H2 antihistamine antagonists has revolutionized the treatment of that disease. A benzimidazole, Omeprazole (55), inhibits gastric secretion and subsequent ulcer formation by a quite different mechanism. Studies at the molecular level suggest that this compound inhibits K+/H+ dependent ATPase and consequently shuts down the proton pumping action of this enzyme system. Treatment of pyridyl carbinol 51 with thionyl chloride leads to the corresponding chloride (52). Treatment of that intermediate with 5-methoxy-2-mercaptobenzimidazole (53), obtained from reaction of 4-methoxy-o-phenylenediamine with potassium ethylxanthate leads to displacement of halogen and formation of the sulfide (54). Finally, oxidation with 3~chloroperbenzoic acid produces the sulfoxide omeprazole (55) [17]. OMc

(51);X = OH

H (53)

(»);x.a

(CA\, y,. .

(ls)|x = b

The rj-fluorobutyrophenone group is one of the hoary traditions in the field of antidopaminergic antipsychotic drugs. First introduced in the neuroleptic haloperidol, this group has appeared in numerous drugs or drugs-to-be. It is of interest that an isoxazole serves as a surrogate for the carbonyl group in some ]>fluorobutyrophenones. Alkylation of l-(4-piperidinyl)-2~ benzimidazolinone (57), an intermediate closely related to one used for domperidone (see Volume 3, p. 174) with 3-(3-chloropropyl)-6-fluoro-l,2-benzisoxazole (56) affords neflumozide (58) [18,19].

(56)

(57)

(58)

tf

134

Five-Membered Ring Benzofused Heterocycles 7. BENZOTHIAZOLE

Calcium channel antagonists have proven of great value as antianginal and antihypertensive agents. Most of these agents fall into one of three rather narrow structural classes. It is thus of interest to find that a structurally quite different benzothiazole shows the same type of activity. It is of note too that the agent in question, fostedil (63), is one of the very few phosphorous containing agents to be developed for the clinic. Treatment of benzanilide 59 with phosphorous pentasulfide or Lawesson's reagent gives thioamide 60. Oxidative ring formation by reaction with potassium ferricyanide and base (presumably involving a free radical intermediate) constructs the benzothiazole ring of 61. Bromination of this compound with N-bromosuccinimide produces bromomethyl intermediate 62. The synthesis of fostedil (63) concludes with a Michaelis-Arbuzov reaction of 62 with triethyl phosphite f20].

c

r

u x (59)

.

-

c

r u (60)

x

. O

(61); X = H (62); X = Br

(63)

Tiaramide (67) is a benzothiazolinone containing antiasthmatic agent. One of its syntheses begins with alkylation of 5~chloro-2-aminobenzothiazole (64) by 4-(2-hydroxyethyl)-l~ piperazinylcarbonylmethyl chloride (65) to give 66 and concludes by gentle hydrolysis with methanolic hydrogen chloride to convert the imino moiety to the carbonyl of tiaramide [21].

v

(64)

(65)

_

7 O H

Y r^ CH2CON N s ^ ^ V—/ OH (66); X = NH (67); X = O


135

REFERENCES 1. A. M. Lefer, Drugs of the Future, 11, 197 (1986). 2. R. A. Johnson, E. G. Nidy, J. W. Aiken, N. J. Crittenden, and R. R. Gorman, J. Med. Chem., 29,1461 (1986). 3. M. A. Bertola, W. J. Quax, B. W. Robertson, A. F. Marx, C. J. Van der Laken, H. S. Koger, G. T. Phillips, and P. D. Watts, Eur. Pat. Appl. 233,656 (1987) via Chem. Abstr., 108: 20,548m (1987). 4. D. Frehel, R. Boigegrain, and J.-P. Maffrand, Heterocycles, 22, 1235 (1984). 5. Anon,, Belg. Pat. 766,392 (1971); via Chem. Abstr. 76: 140,493g (1972). 6. D. L. Lednicer and L. A. Mitscher, The Organic Chemistry of Drug Synthesis, Wiley, New York, 1980, Vol. 2, p. 355. 7. N. Gruenfeld, J. L. Stanton, A. ML Yuan, F. H. Ebetino, L. J. Browne, C. Glide, and C. F. Huebner, J. Med. Chem., 26, 1277 (1983). 8. C. J. Blankley, J. S. Kaltenbronn, D. E. DeJohn, A. Werner, L. R. Bennett, G. Bobowski, U. Krolls, D. R. Johnson, W. M. Pearlman, M. L. Hoefle, A. D. Essenburg, D. M. Cohen, and H. R. Kaplan, /. Med. Chem., 30, 992 (1987). 9. J. R. McCarthy, D. L. Trepanier, and M. E. Letourneau, Eur. Pat. Appl. 127,143 (1984) via Chem. Abstr., 102: 149,101r (1985). 10. G. M. Shutske, L. L. Setescak, R. C. Allen, L. Davis, R. C. Effland, K. Ranbom, J. M. Kitzen, J. C. Wilker, and W. J. Novick, Jr., /. Med. Chem,, 25, 36 (1982). 11. H. Uno, M. Kurokawa, Y. Masuda, and H. Nishimura, J. Med. Chem., 22,180 (1979). 12. H. Uno and M. Kurokawa, Chem. Pharm. Bull., 26, 3498 (1978). 13. R. E. Brown, V. St. Georgiev, and B. Loev, 17.5. Patent 4,298,742 (1981) via Chem. Abstr., 96: 68,972f (1981). 14. T. A. Crowell, U.S. Patent 4,424,362 (1984) via Chem. Abstr., 100: 121,077x (1984). 15. R. J. Gyurik and W. D. Kingsbury, U. S. Patent 4,258,198 (1981) via Chem. Abstr., 95: 7,284r(1981) 16. A. H. M. Raeymalkers and J. H. L. VanGelder, U.S. Patent 4,032,536 (1977) via Chem. Ator.,87:168,033h(1977). 17. P. Lindberg, P. Nordberg, T. Alminger, A. Braudstrom, and B. Wallmark, /. Med. Chem., 29, 1327 (1986). ! 8. U. K. Junggren and S. E. Sjostrand, Eur. Pat. Appl. 5,129 (1979); via Chem. Abstr. 92: 198,392z (1979). 1(). L. Davis and J. T. Klein, U.S. Patent 4,390,544 (1983); via Chem. Abstr. 99: 122,458r (1983).

136


20. K. Yoshino, T. Kohno, T. Uno, T. Morita, and G. Tsukamota, J. Med. Chem., 29, 820 (1986). 21. T. Masuko, Jpn, Kokai 55/145,673 (1980); via Chem. Abstr. 94: 192,376n (1980).

8

Six-Membered Ring

Benzofused Heterocycles

1. CHROMONES The great commercial success of the adrenergic p-receptor blockers has led to the synthesis of many other potential drugs by the incorporation of an oxypropanolamine functionality into a wide variety of aromatic rings. This has lead to the preparation of thousands of such analogues. An example of this is flavodilol (3), an antihypertensive agent, which is synthesized using chemistry standard for the purpose by starting with 7-hydroxyflavone (1). The side chain scaffolding is assembled by reaction with epichlorohydrin to give glycidylether 2 and this is in turn reacted with propylamine to giveflavodilol(3) 11]. Surprisingly, flavodilol does not block the effects of norepinephrine at either adrenergic a or p receptors, despite its formal structural similarity to the classical p-bloekers.

CH2CH2Mc

The attractive properties of cromolyn as an inhibitor of the release of mediators of anaphylaxis has inspired many attempts to improve on the antiasthmatic characteristics of that substance. One such agent is cromitrile (6). In this case, a tetrazolyl unit is introduced as a carboxy group 137

Six-Membered Ring Benzofused Heterocycles

138

bioisostere and the two aromatic ring systems are different from one another. One synthesis concludes in the classic manner by converting chromone ester 4 to the carboxamide (5) by ammonolysis, dehydrating the latter functionality to the nitrile with tosyl chloride in pyridine and adding sodium azide selectively across this functionality to produce the antiasthmatic agent cromitrile (6) [2], The chemical basis for the selectivity of the last reaction is not obvious.

NC (4); X = OEt (5); X = NH2

(6)

2. BENZODIOXANES Azaloxan (12) is an antidepressant agent. Its synthesis can be accomplished starting with the reaction of catechol (7) and 3,4-dibromobutyronitrile (obtained by addition of bromine to the olefin) to give l,4-benzodioxan-2-ylacetonitrile (8). A series of functional group transformations ensues [hydrolysis to the acid (9), reduction to the alcohol (10) and conversion to a tosylate (11)] culminating in an SN-2 displacement reaction on tosylate 11 with l~(4~piperidinyl)-2"-imidazolidinone to give azaloxan (12) [3].

S

OH

(7)

(8); X = CN (9); X = CO2H U0);X = CH2OH (11); X = CH2OTos


139

3. QUINOLINES AND CARBOSTYRILS A partially reduced quinoline derivative with antiulcerative and antisecretory activities is isotiquimide (14). It may be synthesized by metallating (with n-BuLi) 4-methyl-5,6,7,8-tetrahydroquinoline and condensing this with dimethylmethoxysilylisothiocyanate to produce the desired thioamide isotiquimide (14) [4]. N { ^ Me (13) Losulazine (20) is an orally and parenterally active antihypertensive agent apparently acting on peripheral postganglionic sympathetic nerve terminals to deplete norcpinephrine stores. This prevents vasoconstriction by reducing levels of that neurotransmitter available following stimulation of adrenergic nerves. In this sense its calming action mimics that of reserpine. In a convergent synthesis, 4-nitrobenzoyl chloride (15) is used to monoacylate piperazine to give amide 16; the nitro moiety of this is then reduced to give substituted aniline synthon 17. A nucleophilic aromatic displacement reaction between 17 and 4~chloro-7-trifluoromethylquinoline (18) leads to 4-aminoquinoline derivative 19. The synthesis of losulazine concludes by formation of the sulfonamide (20) by reaction of the remaining secondary piperazine nitrogen with 4-fluorobenzenesulfonyl chloride [5J.

(16); X = 2 (17); X = NH2

(18)

CON NR


140

The continuing development of resistant strains by the malaria parasite to chemotherapeutic agents has led to an ongoing, though sporadic, synthesis of new chemical entities aimed at that target in the hope of finding drugs which will overcome resistance. That there is still interest in the classical 4-aminoquinolines (see The Organic Chemistry of Drug Synthesis, Volume 1, p. 343) is attested to by the preparation of tebuquine (23), a hybrid compound combining structural elements of amodiaquine and some 2-(dialkylamino)-o~cresols which have shown antimalarial activity. The synthesis is closely analogous to the original method used for amodiaquine. It starts by reacting 4'-chloro-N-(6-hydroxyl[l,l'-biphenyl]-3-yl) acetamide (21) with formaldehyde and tbutylamine in an aromatic version of the Mannich reaction to give acetanilide 22. Hydrolysis to the free aniline with HC1 is followed by the standard nucleophilic aromatic displacement reaction with 4,7-dichloroquinoline to complete the synthesis of the antimalarial agent tebuquine (23) |6].

CH 2 NUCMc 3 OH

ACN:

AcNI(21)

(22)

P r o c a t e r o l (27) is a p - a d r e n e r g i c a g o n i s t w i t h b r o n c h o d i l a t o r y action i n t e n d e d for u s e in b r o n c h i a l a s t h m a . A c a r b o s t y r i l N H g r o u p i n t h i s m o l e c u l e is u s e d a s p a r t o f a b i o i s o s t e r e r e p l a c i n g t h e c a t e c h o l m o i e t y o f e p i n e p h r i n e . P h a r m a c o l o g i c a l l y it i s o f s p e c i a l i n t e r e s t i n t h a t it i s s a i d t o selectively stimulate adrenergic P-2 receptors w i t h o u t m u c h effect o n p-1 receptors, m i n i m i z i n g cardiac stimulation. Friedel-Crafts reaction between 8-hydroxycarbostyril (24) and 2b r o m o b u t y r y l b r o m i d e l e a d s t o t h e e x p e c t e d a c y l a t i o n at C - 5 i n t h e m o r e a c t i v a t e d ring ( 2 5 ) . H a l i d e d i s p l a c e m e n t w i t h i s o p r o p y l a m i n e (to g i v e 2 6 ) is f o l l o w e d b y b o r o h y d r i d e r e d u c t i o n to the m i x t u r e o f d i a s t e r e o i s o m e r i c a r y l e t h a n o l a m i n e s . T h e e r y t h r o - i s o m e r is p r o c a t e r o l ( 2 7 ) [ 7 ] .


141 NHCHMe2

OH (24)

OH (26)

4. Q U I N O L O N E S Intense synthetic activity has resulted centered about congeners of the quinolone-3-carboxylic acids w h e n it was found that certain analogues, notably those with 6-fluoro~7~piperazinyl moieties (now k n o w n collectively as the fluoroquinolones) possess broad spectrum oral antimicrobial activity with potencies in the range of fermentation-derived antibiotics. Additional interest in these molecules was inspired by the discovery of their unusual molecular m o d e of action; strongly selective inhibition of the action of bacterial D N A gyrase, a type II (double strand D N A breaking) topoisomerase essential for dictating the conformation of bacterial circular D N A . The activity of D N A gyrase is essential for the orderly processing of D N A as several important enzymes are sensitive to the topological arrangement of the molecule. Several thousand furoquinolone analogues have been prepared and several (including norfloxacin, ciprofloxacin, and ofloxacin) have been introduced clinically and several others are advancing toward commercial use. Pefloxacin (33) is the N-methyl analogue of norfloxacin (58) and is at least partly converted to it by metabolic e n z y m e s in vivo. It has been launched in France for the treatment of a n u m b e r of infections including those caused by sensitive strains of Pseudomonas

aeruginosa.

It can be syn-

thesized starting with the Gould-Jacobs reaction of 3-chloro-4-fluoroaniline (28) and diethyl ethoxymethylenemalonate in an addition-elimination sequence leading to 29 which undergoes


142

thermal cyclization to hydroxyquinoline ester 30. Despite the apparent asymmetric substitution pattern of 29 which might lead one to believe that mixtures would result from the pericyclic reaction, only one product is observed in any quantity. It is rationalized that the buttressing effect of the aryl chlorine atom is the cause of the observed steric preference. Next, alkylation on nitrogen with EtI leads to quinolone ester 31 which is itself readily hydrolyzed to the corresponding carboxylic acid 32. A nucleophilic aromatic displacement reaction of the 7-chloro moiety, activated by the pyridone carbonyl as a sort of vinylogous acid chloride, with N-methylpiperazine completes a synthesis of pefloxacin (33). This synthetic sequence is the classic route to this series of antimicrobial agents [8]. COÊl

(31), RsEt (32), R=H

(33)

O n e of the most promising newer members of the fluoroquinolone family is ofloxacin (40). In this case, the N~ethyl moiety has been m a d e rigid by incorporation into a heterocyclic ring. This creates a chiral center and subsequent chiral synthesis reveals that the S-enantiomer is significantly m o r e potent than its antipode and is rather more water soluble than the racemate. Of the various syntheses of ofloxacin, the more recent chiral process is illustrated. Starting with ethyl (2,3,4,5tetrafluoro)-3-oxopropionate (34), the ethoxymethylene function is introduced by cross condensation with ethylorthoformate and acetic anhydride to give 3 5 . Alaninol, prepared by reduction of optically active alanine, is incorporated in an addition-elimination reaction to give 36. Treatment of the latter with non-nucleophilic base leads to a nucleophilic aromatic displacement of fluorine and thus ring closure to give 37. T h e geometry of 3 6 appears to be unimportant as it is lost in the intermediate anion. Condensation, of course, depends upon proximity. Repetition of the base


143

treatment and/or use of forcing conditions leads to a second displacement of aromatic fluorine and the closure of the second heterocyclic ring of 38. The remainder of the sequence follows the standard quinolone synthetic route in which saponification (to 39) and nucleophilic aromatic displacement with N-methylpiperazine leads to ofloxacin (40). The chirality of the alanine derivative employed dictates the stereochemistry of the final product [9,10,11].

(34)

(35)

U

F XÔH Me (37>

-^Mc

(38);Rs=Et (39);R = H

(36)

McN^J

VMe

(40)

JJ ~

Et (41)

Norfloxacin (41), the substance which triggered this avalanche of activity, has recently been introduced into clinical practice in the United States. Its synthesis parallels closely that of its Nmethyl analogue, pefloxacin, except that the nucleophilic aromatic displacement reaction of 32 is carried out with mono-N-carboethoxypiperazine instead and the final step encompasses deblocking of this carbamoyl ester moiety [8]. The generally accepted structure-activity relationships developed in the early work in the quinolone series held that the N-l substituent needed to be small and aliphatic. This picture was upset in a dramatic way with the discovery of the excellent potency and antimicrobial spectrum of difloxacin (45) and its congeners in which the substituent on N-l is an aromatic ring. The synthe-


144

sis of difloxacin begins with 2,4-dichloro-5-fluoroacetophenone (42) which is carbethoxylated by reaction with diethyl carbonate and base to give ester 43. The rest of the synthesis parallels that described above for ofloxacin except that the addition-elimination reaction is carried out with 4fluoroaniline and 44 is the resulting product after cyclization and hydrolysis. Conclusion of the synthesis of difloxacin (45) involves displacement of the activated 7-chloro atom by Nmethylpiperazine [12]. CCXH

(44)

(45) Amifloxacin (48) demonstrates that the methylene group of the N-ethyl substituent can be replaced successfully by a bioisosteric N H group in the fluoroquinolone series. In one of two interesting syntheses, 4-hydroxyquinoline ester 4 6 is reacted with O-(2,4-dinitrophenyl)hydroxylamine with the aid of potassium carbonate to give the hydrazine analogue 47. Next, N-formylation is accomplished with aceticformic anhydride (formed in situ by reaction of A c 2 O + H C O 2 H ) . T h e formamide thus produced is methylated with iodomethane and base. This sequence allows for clean monoalkylation. T h e N-formyl group is hydrolyzed with N a O H and N-methylpiperazine displacement of the activated C-7 chloro atom completes the synthesis of amifloxacin (48) [13].

o COoEt

CO2Et

(46)

(47)

CO2H

(48)


145

Alternately, amifloxacin can be prepared via the ofloxacin/difloxacin route using an additionelimination reaction with unsymmetrical N-methyl-N-formyl hydrazone to give 49 [14].

(48)

Enoxacin (58) is an analogue of the quinolones based on the 1,8-naphthyridine ring system. The drug is available commercially in Japan. The naphthyridine analogues of the fluoroquinolones, generally speaking, are not as potent as the quinolones in vitro but have favorable pharmacokinetic characteristics which help compensate for this in vivo. The synthesis of enoxacin begins by reaction of 2,6-dichloro-3-nitropyridine (50) with N-carboethoxypiperazine to give 51. The second, less reactive, halo group is then displaced with ammonia, and the resulting amine acetylated to 52, the nitro group is then reduced to give triaminopyridine derivative 53. Diazotization to 54 with sodium nitrite plus tetrafluoroboric acid, followed by heating in xylene, results in displacement of the diazonium salt by fluorine to give 55. The rest of the synthesis follows the well trodden path of acid hydrolysis followed by Gould-Jacobs reaction to 56. The alternate ortho-closure onto pyridine nitrogen does not take place presumably due to the steric interference by the piperazinyl group at C-7. Sequential hydrolyses to 57 and then 58 concludes the synthesis of enoxacin [15,16|. O2R EtOCON^ (50)

EtOCON^ 51

(52);X = O (53);X = H OH

(54); X = N2 (55);X = F

(56)

(57); X = CO2Et (58);X = H


146 5. TETRAHYDROISOQUINOLINES

Most of the widely used antidepressants are tricyclics related to imipramine. A 1-phenyltetrahydroisoquinoline analogue, nomifensine (60), departs from this structural pattern. Pharmacologically it inhibits the reuptake of catecholamines such as dopamine at neurons. It can be synthesized by alkylation of 2-nitrobenzyl-methylamine with phenacyl bromide followed by catalytic reduction of the nitro group (Pd-C) and then hydride reduction of the keto moiety to give 59. Strong acid treatment leads to cyclodehydration to nomifensine (60) [17]. NH-

NH-

NMe

N ' Mc

(59)

(60)

Given the well-established structural forgiveness available among angiotensin-converting enzyme (ACE) inhibitors, it is not surprising to find a tetrahydroisoquinoline analogue represented. Quinapril (64) is synthesized from the t-butyl ester of tetrahydroisoquinoline-3-carboxylic acid (61) by amide formation with 62 using mixed ester methodology. Subsequent partial deblocking of 63 upon acid treatment leads to quinapril (64) [18]. This is by now a familiar reaction sequence for construction of molecules in this class. J»CO2CMe3

(61)

,-CO2El (62)

(63);R (64); R = H

6. BENZAZEPINES Trepipam (69) is a sedative agent apparently acting via dopaminergic mechanisms. It can be synthesized by attack on the less hindered terminus of styrene oxide (66) by 4,5-dimethoxyphenethylamine (65) to give 67. Cyclodehydration catalyzed by strong acid then leads to 68 and N-


147

methylation with formic acid and formaldehyde (Eschweiler-Clarke reaction) completes the synthesis of trepipam (69) [19].

(65)

(66)

(67)

(68);X = (69); X = Me

Fenoldopam (76) is an antihypertensive renal vasodilator apparently operating through the dopamine system. It is conceptually similar to trepipam. Fenoldopam is superior to dopamine itself because of its oral activity and selectivity for dopamine D-1 receptors (D-2 receptors are associated with emesis). It is synthesized by reduction of 3,4-dimethoxyphenylacetonitrile (70) to dimethoxyphenethylamine (71). Attack of this last on 4-methoxystyrene oxide (72) leads to the product of attack on the epoxide on the less hindered side (73). Ring closure with strong acid leads to substituted benzazepine 74. O-Deaikylation is accomplished with boron tribromide and the catechol moiety is oxidized to the ortho-quinone 75. Treatment with 9N HC1 results in conjugate (1,6) chloride addition and the formation of fenoldopam (76) [20,21].

OH (76)


148 7. BENZOTHIEPINS

Enolicam (81) is a nonsteroidal antiinflammatory agent which may be viewed as a higher homologue of compounds in the pyroxicam series (see The Organic Chemistry of Drug Synthesis, Volume 2, p. 394). It is intended for use in the treatment of psoriasis and arthritis. It is active both orally and topically. Oxidation of 7-chlorotetrahydro-l-benzothiepen-5-one (77) with hydrogen peroxide leads to the sulfone 78. This is converted to the pyrrolidine enamine (79) using tosic acid as catalyst and this is then reacted at its electron rich center with 3,4-dichlorophenylisocyan~ ate to give amide 80. The synthesis concludes with hydrolysis of the enamine function with HC1 to produce enolicam (81). Enolicam is sufficiently acidic to be used primarily as its sodium salt [22].

(80)

(81)

8. QUINAZOLINES AND QUINAZOLINONES Doxazosin (84) is an adrenergic postsynaptic oc-1 receptor antagonist with antihypertensive properties. The discerning eye will recognize a structural resemblance to the antihypertensive quinazoline prazosin, also an a-1 receptor antagonist. Doxazosin was produced in an attempt to develop agents which could be administered once daily to combat hypertension. It is synthesized by reacting 4-amino-2-chloro-6,7-dimethoxyquinazoline (82) with (l,4-benzodioxan-2-ylcarbonyl)piperazine (83) in an addition-elimination sequence leading to 84 [23].


149

J

V

O

I

M< (82)

(84)

(83)

Trimetrexate (88) is an antineoplastic agent related to the well-established folic acid antimetabolite methotrexate. It can be synthesized by selective diazotization of the most basic ami no group of 2,4,6-triamino~5-methylquinazoline (85) followed by a Sandmeyer displacement with C u C N to give nitrile 8 6 . Careful reduction using Raney nickel produces the aminomethyl intermediate 87 or, if the reaction is carried out in the presence of 3,4,5-trimethoxyaniline, t r i m e t r e x a t e (88) [24]. O n e presumes that that outcome is a consequence of amine exchange at the partially reduced i m i n e stage and further reduction.

HUN OMc (88)

Alfuzosin (91) is a p r a z o s i n - l i k e hypotensive adrenergic ot-1 receptor blocker with the special structural feature that two carbons have been excised conceptually from the piperazine ring normally present in this series. Following the usual sequence for this series, reaction of 4-amino-2chloro-7-dimethoxyquinazoline (89) with the tetrahydro-2-furyl amide of 3-methylaminopropylamine (90) gives alfuzosin (91) [25]. Alfuzosin is claimed to cause less orthostatic hypotention (dizziness or fainting upon sudden rising) than prazosin.

150


MeNHCH2CH2CH2NHCO—£ > (89)

2

-NC NH2 (91)

The quinazolinone-containing antiallergic agent tiacrilast (95) has many of the pharmacological properties of the mediator release inhibitor cromolyn sodium for treatment of bronchoconstriction and seems in animal models to be orally active as well. Its structure embodies a molecular simplification (molecular dissection strategm) of a lead series of pyrido[2,l-b|quinazolinecarboxylic acids. It is synthesized in straightforward fashion by reacting 5-methylthioanthranilic acid (92) with formamide to form the quinoxalone ring of 93. This is then subjected to an additionelimination reaction with E~3-chloroacrylate to give methyl ester 94 which itself is then hydrolyzed to produce tiacrilast (95) [26],

(92)

(93)

(94);R = Me (95); R = H

Fluproquazone (97) contains a 2-quinazolinone nucleus and is found to be an analgetic agent useful in mild to moderate pain. One of the preparations involves reaction of 2-isopropylamino-4methyl-4'-fluoro-benzophenone (96) with potassium cyanate in hot acetic acid [27].


151

Altanserin (100) is a representative of the thiaquinazolinones. This serotonin antagonist is said to prevent gastric lesions. One method for preparation of this compound involves first preparation of isothiocyanate derivative 99, by reacting 4-fluorobenzoylpiperidine with 2bromoethylamine and then converting the intermediate to the isothiocyanate with thionyl chloride and base. Condensation of 99 with methyl anthranilate (98) probably proceeds initially to a thiourea. Cyclization by ester-amide interchange leads to altanserin (100) [28].

S=C=NCH2CH2N~y— COs

C02Me (98)

(99)

T /\ fQ-c ,NCH2CH2NV~CO (100)

9. PHTHALAZINES Oxagrelate (104) is of interest as a platelet antiaggretory agent and is thus of potential value in preventing thrombus formation in blood vessels. It may also be of potential value in preventing arteriosclerotic lesions in coronary arteries - a substantial cause of morbidity and mortality in


152

western countries. It is synthesized by reacting 4-ethoxycarbonyl~5-methylphthalic anhydride (101) (itself derived from the Diels-Alder product of dimethylacetylenedicarboxylate and ethyl isodehydroacetate) with excess malonic acid in pyridine to give phthalide 102; the reaction apparently proceeds by attack of malonate on the more electrophilic carbonyl group. The methyl group introduced into 101 is the remnant of the malonic acid moiety after decarboxylation. Oxidation of of the newly introduced methyl group of 102 with aqueous permanganate produces the keto acid; this is converted to the phthalazine carboxylic acid with hydrazine and that esterified to 103. Treatment of this last with sodium borohydride completes the synthesis of oxagrelate [29]. CO2Et

(101)

(102)

(103)

(104) Azelastine (107) is an antiallergic/antiasthmatic agent prepared from 4~chlorobenzyl~2'-carboxyphenylketone (105) by condensation with hydrazine to give the phthalazinone (106) followed by reaction of the sodium salt of this last with 2-(2-chloroethyl)-N-methylpyrrolidine (presumably involving nucleophilic ring expansion of the bicyclic quaternary salt putatively formed as a first product) to complete the synthesis [30].

(105)

(106)

(107)


153

10. BENZODIAZEPINES That once well-represented class of compounds, the benzodiazepine anxiolytic agents, has declined precipitously in numbers in consecutive volumes of this series. Preparation of the sole classical representative in the present volume starts with the preformed 7~chloro-5-(ochlorophenyl)-2,3-dihydro~lH~l,4-benzodiazepine (108) (see The Organic Chemistry of Drug Synthesis, Volume 1, page 369). Condensation of that with 2-chloroacetylisocyanate proceeds on the more basic nitrogen of 108 to afford urea 109. Reaction of that with sodium iodide and base probably proceeds initially by halogen exchange of iodine for chlorine (Finkelstein reaction). Subsequent replacement of iodide by the enol anion of the urea oxygen results in formation of the oxazolone ring. There is thus obtained reclazepam (110) [31].

(108)

(109)

(110)

REFERENCES 1. E. S. C. Wu, Eur. Pat. Appl, 81,621 (1983) via Chem. Abstr., 99: 139,772r (1983). 2. J. W. Spicer and B. T. Warren, U. S. Patent 4,238,495 (1980) via Chem. Abstr., 94: 139,623q (1980). 3. C. F. Huebner, U. S. Patent 4,329,348 (1982) via Chem. Abstr., 97: 92,295d (1982). 4. R. Crossley, Brit. Pat. Appl. 2,122,615 (1984) via Chem. Abstr., 100: 191,753p (1984). 5. J. M. McCall, U. S. Patent 4,167,567 (1979) via Chem. Abstr., 92: 6,555f (1980). 6. L. M. Werbel, P. D. Cook, E. F. Elslager, J. H. Hung, J. L. Johnson, S. J. Kesten, D. J. McNamara, D. F. Ortwein, and D. F. Worth, J. Med. Chem., 29, 924 (1986).

154


7. S. Yoshizaki, K. Tanimura, S. Tamada, Y. Yabuuchi, and K. Nakagawa, /. Med. Chem., 79,1138(1976). 8. H. Koga, A. Itoh, S. Murayama, S. Suzue, and T. Irikura, /. Med. Chem., 23,1358 (1980). 9. H. Egawa, T. Miyamoto, and J.-I. Matsumoto, Chem. Pharm. Bull, 34,4098 (1986). 10. I. Hayakowa, T. Hiramitsu, and Y. Tanaka, Chem. Pharm. Bull, 32,4907 (1984). 1L L. A. Mitscher, P. N. Sharma, L. L. Shen, and D. T. W. Chu, /. Med. Chem., 31, 2283 (1988). 12. D. T. W. Chu, P. B. Fernandes, A. K. Claiborne, E. Pihuleac, C. W. Nordeen, R. E. Maleczke, Jr., and A. G. Pernet,./. Med. Chem., 28, 1558 (1985). 13. M. P. Wentland, D. M. Bailey, J. B, Cornett, R. A. Dobson, R. G. Powles, and R. B. Wagner, J. Med. Chem., 27,1103 (1984). 14. D. T. W. Chu, /. Heterocycl. Chem., 22, 1033 (1985). 15. J.-I. Matsumoto, T. Miyamoto, A. Minamida, Y. Nishimura, H. Egawa, and H. Nishimura, J. Heterocycl. Chem., 21, 673 (1984). 16. J.-I Matsumoto, T. Miyamoto, A. Minamida, Y. Nishimura, H. Egawa, and H. Nishimura, J. Med. Chem., 27, 292 (1984). 17. E. Zara-Kaczian, L. Gyorgy, G. Deak, A. Seregi, and M. Doda, /. Med. Chem., 29, 1189 (1986). 18. S. Klutchko, C. J. Blankley, R. W. Fleming, J. M. Hinkley, A. E. Werner, I. Nordin, A. Holmes, M. L, Hoefle, D. M. Cohen, A. D. Essenburg, 'and H. R. Kaplan, J. Med. Chem., 29, 1953 (1986). 19. C. Kaiser, P. A. Dandridge, E. Garvey, R. A. Hahn, H. M. Sarau, P. E. Setler, L. S. Bass and J. Clardy, J. Med. Chem., 25, 697 (1982). 20. J. Weinstock, D. L. Ladd, J. W. Wilson, C. K. Brush, N. C. F. Yin, G. Gallagher, Jr., M. E. McCarthy, J. Silvestri, H. M. Saran, K. E. Flaim, D. M. Ackerman, P. E. Setler, A. J. Tobia, and R. A. Hahn, /. Med. Chem., 29, 2315 (1986). 21. J. Weinstock, J. W. Wilson, D. L. Ladd, C. K. Brush, F. R. Pfeiffer, G. Y. Juo, K. G. Holden, N. C. F. Yin, R. A. Hahn, J. R. Wardell, Jr., A. J. Tobia, P. E. Setler, H. M. Saran, and P. T. Ridley, J. Med. Chem., 23,973 (1980). 22. M. H. Rosen, US. Patent 4,185,109 (1980) via Chem. Abstr., 93:46,451w (1980). 23. S. F. Campbell, M. J. Davey, J. D. Hardstone, B. N. Lewis, and M. J. Palmer, /. Med. Chem., 30,49 (1987). 24. E. F. Elslager and L. M. Werbel, Brit. Pat. Appl. 1,345,502 (1973) via Chem. Abstr., 80: 133,461z (1974). 25. P. M. Manoury, J. L. Binet, A. P. Damas, F. Lefevre-Borg, and I. Cavero, J. Med. Chem., 29, 19 (1986).


155

26. R. A. LeMahieu, M. Carson, W. C. Nason, D. R. Paixish, A. F. Welton, H. W. Baruth, and B. Yaremko, /. Med. Chem., 26,420 (1983). 27. G. E. Hardtmann, US. Patent 3,937,705 (1976) via Chem. Abstr., 84: 135,716t (1976). 28. V. Aguirre Ormaza, Span. 548,966 (1986) via Chem. Abstr., 106: 84,630y (1986). 29. M. Ishikawa, Y. Eguchi, and A. Sugimoto, Chem. Pharm. Bull, 28, 2770 (1980). 30. K. Tasaka and M. Akagi, Arzneim.-Forsch., 29,488 (1979). 31. P. K. Yonan, US. Patent 4,208,327 (1980) via Chem. Abstr., 94: 30,804y (1980).

9

Bicyclic Fused

Heterocycles

A large number of benzo-fused heterocyclic nuclei are of course possible in theory. This number is, however, dwarfed by the structural possibilities brought into play by fusing two heterocyclic rings. A large number of such structures have in fact been synthesized in the search for therapeutic agents. Part of the impetus probably comes from the fact that half the bases involved in genetic material, the purines, in fact consist of fused heterocycles, specifically imidazo[l,2-a]~ pyrimidines. Another motivation comes from the fact that the large number of structural possibilities opens the way for good patent exclusivity. 1. INDOLIZINES Aminoalkyl ethers of 3-benzoyl benzofurans such as amiodarone have been found to be very effective antiarrhythmie agents. Biological activity is retained when the heterocyclic nucleus is replaced by the nearly isosteric indolizine system. Reaction of 2-picoline (1) with ethyl chloroacetate leads to the acyl pyridinium salt 2; reaction of that with propionic anhydride leads to the indolizine 4, possibly via intermediate 3 (Chichibabin synthesis). Friedel-Crafts acylation of 4 with the 2-toluenesulfonate ester of rj-hydroxybenzoyl chloride gives the ketone 5. The rj-toluenesulfonate protecting group is then removed by saponification. Treatment of the resulting phenol 6 with 1,3dibromopropane and dibutylamine gives the antiarrhythmie agent butoprozine (7) [1]. 2. PYRROLIZINES Incorporation of the 2-aryl-2-methylacetic acid moiety characteristic of NSAID's as part of 157

Bicyclic Fused Heterocycles

158

a fused heterocyclic ring is consistent with good activity. A number of syntheses have been described for compounds in the anirolac series [2,3,4]. One of the more interesting preparations starts by acylation of 2-(methylthio)pyrrole 8 with anisoyl chloride (9). Oxidation of the product 10 with peracid leads to the sulfone 11. The nitrogen anion obtained from treatment of 11 with

Me V

(1)

B

o

(2)

OCH2C H2CH2N(C4H9)2 (7)

(5);R==u-CH3C6H4SO2 (6);R = H

(4)

base is then allowed to react with the Meldrum ester 12. Attack of the anion on the cyclopropyl carbon leads to ring opening and formation of the alkylation product 13. It was found empirically that the next step proceeds in better yield with a methyl ester; this intermediate (14) is obtained by ester interchange with methanol. T h e anion from malonate 14 then displaces the pyrrole sulfone group to give the cyclization product 15. Saponification, followed by decarboxylation of the resulting diacid, affords anirolac (16) [3,4].

3. C Y C L O P E N T A P Y R R O L E S T h e majority of e n d o g e n o u s prostaglandins tend to exert undesirable effects on the cardiovascular system. T h e s e c o m p o u n d s as a rule tend to cause vasoconstriction and promote platelet aggrega-


159

tion. An important exception to this trend is the last prostaglandin derivative to be discovered, PGI2, or prostacyclin, (epoprostenol, 24). Use of this compound in therapy is severely limited by its short biological half-life, which is measured in minutes. It has been found that compounds in which the ring containing the enol ether function is replaced by some more stable ring system retain a considerable portion of the activity of the endogenous compound. The starting material for an analogue in which that ring is replaced by a pyrrole, PGF2a (17), is interestingly the natural precursor for 24 as well. Treatment of PGF2a methyl ester (18) with base and iodine, gives COCl ^N>-SMe + H (8)

CH2CH2CH(CO2Mc)2 (14) (13)

McO CO2Me

C02H

CO2Me (15)

(16)

the iodolactonization product 19. Regiochemistry is controlled by the close approach which is


160

possible between the ring hydroxyl and the isolated olefin. The remaining free hydroxyl groups are then protected as their tetrahydropyranyl ethers (20). Dehydroiodination with DBN, followed by hydrolysis and then by oxidation of the C-9 hydroxyl group gives diketone 21. Condensation of 1,4 diketones with secondary amines is one of the classical methods for forming pyrroles. Reaction of 21 with aniline thus proceeds to give the highly substituted pyrrole 22. Saponification of the ester and removal of the tetrahydropyranyl groups completes the preparation of the antiasthmatic agent piriprost (23) [5].

CO2Mc

THPd

6THP

6THP

(21)

(22)

H0 2 C

6H

(23)

(24)


161

4. IMIDAZOPYRIDINES The continuing search for effective platelet aggregation inhibitors has covered a wide variety of structural types. One such candidate consists of a fatty acid chain attached to a fused heterocycle. (A somewhat fanciful relation to 24 can be imagined.) Reduction of cyanopicoline 25 leads to the primary amine 26; treatment of the corresponding formamide 27, with phosphorous oxychlonde results in cyclodehydration and formation of the imidazo[l,5-a]pyridine 28. The pendant methyl group readily forms an organometallic derivative with n-butyllithium. Treatment of that with the orthoester from 6-bromohexanoic acid gives the alkylation product 30. Hydrolytic removal of the orthoester grouping, leads to pirmagrel (31) [61.

.CN

CH2NH2

Me

Me

(25)

(26)

CH2NHCH Me

Me (27)

CH2(CH2)3CH2C(OEt)3 (30)

(28)

CH2Li (29)

T h e derivative from an isomeric fused system has been described as a sedative-hypnotic c o m p o u n d . T h e synthesis starts by condensation of the aminopicoline 32 with the haloketone 33. T h e resulting pyrrolo[l,2-a]pyridine 34 then undergoes a Mannich reaction with formaldehyde and dimethylamine to give the aminomethylated derivative 3 5 . After quaternization of the dim e t h y l a m i n o group in 3 5 with methyl iodide, the a m m o n i u m group is displaced by cyanide to

162


produce 36. Standard conversion of the nitrile to the amide concludes the synthesis of zolpidem (37) [7]. Selected modified pyrimidines which are closely related to nucleotides have, as noted in Chapter 6, shown therapeutic utility as antiviral and/or antineoplastic agents. Much the same rationale has been used to support the synthesis of compounds related to the purine nucleotides. An analogue of guanine which lacks one of the ring nitrogens, dezaguanine, (47) is, for example, used as an antineoplastic agent. Nitrosation of dimethyl acetone dicarboxylate (38) with nitrous acid gives the derivative 39. This is then reduced to the amine 40. Treatment of the latter with potassium thiocyanate leads initially to addition of the thiocyanate group to the amine to give the thiourea 41. That newly formed function, or more likely an intermediate toward its formation,

"xx

NH 2

(32)

(33)

(35) Me (37)

(34)

(36)


163

condenses intramolecularly with the ketone. There is thus obtained the imidazolethione 42. Ammonolysis proceeds preferentially at the ester on the longer chain to give 43. Treatment of 43 with Raney nickel gives imidazole 44. Exposure of that compound to phosphorus oxychloride serves to convert the amide to the nitrile 45. Exposure of the ester-nitrile to liquid ammonia leads to the amide 46, which is cyclized with sodium carbonate. There is thus obtained dezaguanine, (47) [8]. The narrow therapeutic range of digitalis related cardiotonic agents has resulted in an extensive effort to identify compounds in other structural classes which will improve cardiac function. The discovery of the heterocyclic cardiotonic drug, amrinone, led to research on other heterocyclic compounds for that indication. The imidazopyridine, isomazole (57), is representaCH2CO2Mc ^CH2CO2Mc

(38)

CH2CO2Me

CH2CO2Mc 0=< CHCO2Mc

^ CHCO2Me

NO

NH 2

(39)

(40)

CH2CO2Me

H2NCNH S (44)

(41)

MeO2< (45)

(46)

(47)


164

tive of some of the newer active compounds. Alkylation of the phenolic ester derivative 48 proceeds selectively at the least hindered phenol to give the monobenzyl ether 49. Methylation of the free hydroxyl group (50), followed by removal of the benzyl group by hydrogenolysis gives the phenol 51. This intermediate is then acylated with dimethylthioforrnamidoyl chloride to afford the thiocarbamate 52. This compound, on heating, undergoes an O to § migration of the aryl group to afford the product of replacement of aromatic oxygen by sulfur (53). Removal of the acyl group with aqueous base (54) is followed by methylation of both the thiophenol group and the benzoic acid. Hydrolysis of the methyl ester then gives the benzoic acid 55. Reaction of the carboxyl function with the amino groups in 56, leads to the formation of an imidazole ring. Oxidation of the sulfide moiety to the sulfoxide by MCPBA at low temperature gives isomazole (57) |9]. CO2Me

(48);R = H (49);R = CH2C6H5

CO2Me

CO>Mc

Mc<X

(50); R = CH2C6H5 (51);RSH

CO2H

MeO.

CO2Me

NH, NH, (56) O II SMe

(57)


165

5. PURINEDIONES The purinedione, theophylline (64) has been the mainstay for the treatment of asthma for the better part of a century. Few other drugs rival theophylline for its efficacy as a bronchodilating agent. This drug is, however, noted for its very narrow therapeutic window. Blood levels associated with efficacy range between 10 and 20 mcg/mL; toxic effects start to be seen when those blood levels exceed 25 mcg/mL. A considerable amount of research has thus been devoted to preparing analogues in the hopes of developing a safer agent. The synthesis of one such agent involves a variation on the classical method for the construction of purinones. Condensation of n-propylurea 58 with cyanoacetic acid gives product 59. Treatment of that intermediate with aqueous base leads to addition of urea nitrogen to the nitrile group and consequent formation of the aminouracil 60. Treatment with nitrous acid leads to nitrosation at the only open position on the ring (61); reduction of the newly introduced nitroso group leads to the 1,2 diamine 62. The required remaining carbon atom is then introduced by reaction with formic acid followed by cyclization with sodium hydroxide; there is thus obtained enprofylline (63) [10]. A somewhat more complex theophylline derivative includes both the purinone nucleus and a piperazine side chain more commonly associated with HI antihistaminic compounds. The starting epoxide, 66, is available from treatment of the anion of purinone 65 with epichlorohydrin. Alkylation of the epoxide with monosubstituted piperazine derivative 67, leads to tazifylline (68) [11]. 6. PURINES The deoxyguanine analogue of acyclovir, apparently retains the antiviral activity of the parent compound. Preparation of that agent starts by acylation of the amino group on purine 69. Treatment of that with diacetate 71 under typical glycosidation conditions leads to displacement of the reactive acetal acetate by imidazole nitrogen and the formation of the intermediate 72; removal of the acetate protecting groups with aqueous base affords desciclovir (73) [12]. Preparation of the coccidiostat arprinocid (80) starts by protection of two of the three amino groups on the pyrimidine 74; thus, reaction of 74 with thionyl chloride leads to reaction of

166 O

HN

H-C3H7NHCNH2 + HOCCH2CN (58)

Me (64) Me]

NH nC -H 37 (59)

HNT EC -IH 37 (63) H N NC ( HS 3)2~ (67)

Bicyclic Fused Heterocycels O

k

n-c3H7 (60)

oVNR HN N NH

2 n c H 3 7 (61); R a O (62); R = H2

O H C H C H H N C (H S3)-2 2C 2 N M e (68)

(65);R = (66)R

M C cO C H C H O C H O C O M e (71)

XX) R N H R N H |CH 222 2 R O C H C H O 2 2 2C ((7609));; R R= =» C H ((7723));; R = OMc R~H OMe adjacent amn io groups wtih the bd iendate reagent to ofrm a fused thiadiazole rn ig 75. symmerty of the starting materail guarantees regiochemistry. Reactoin of the product wtih benzyalmn ie 76 gives product 77 by apparent dsipalcement of an amn io group; it is ve that the reaction in fact involves addtion of benzyalmn ie to the pyrm id in ie 6 position, fo loss of ammonai rfom the same center. Acyaltoin of the newyl n itroduced ntirogen gives formamd ie 78. Desulfurization of this last compound wtih Raney nickel wil aford initialy transient formy-ldaimn ie 79. Cyclization of the ofrmamd ie wtih the adjacent amn io group


167 ^F

NH2

NH-

"CH 2 Cl

NH

(74) (77)

Cl

Cl

(80)

(79)

N—CHO

(78)

formation of an imidazole ring; there is thus obtained a r p r i n o c i d (80) [13]. T h e fluorinated derivative of an adenine arabinose glycoside, has shown useful antineoplastic activity. The requisite heterocyclic nucleus is obtained in classic fashion by reaction of the peraminated pyrimidine 8 1 with formamide. In this case too, the symmetry of the starting material means that only a single product, 82, is possible. The remaining free arnino groups are then protected by acetylation (83). Glycosidation of the diacyl compound with the protected chlorinated arabinose derivative 84 affords the protected nucleoside 8 5 . The free diamine 86 is obtained on saponification. Diazotization of this intermediate in the presence of fluoroboric acid in T H F gives the fluorinated derivative 87. T h e regiochemistry of that reaction can probably be attributed to the greater reactivity of the amino group at the 2-position as a consequence of the higher electron density and basicity at that position compared to the 4-amino group. Treatment of the fluorination product with boron trichloride leads to removal of the benzyl protecting groups and the formation offludarabine(88)[14].

168

Bicyclic Fused Heterocycles Bz( M

^ H

(81)

NHR

OBz (85);Rr:COMc (86); R = H

OBz (84);Bz = C6H5CH2

(82);R = H (83);R^COMe

NH2

NH2

OBz

OH

87

(88)

( )

7. TRIAZOLOPYRIMIDINES Inclusion of yet another nitrogen in the aromatic nucleus gives a product, bemitradine (95), described as a diuretic agent. The starting pyrimidine 90 is obtainable in straightforward fashion from condensation of the p-keto ester 89 with guanidine. After protection of 90 as its N-formyl derivative 91, chlorination with phosphorous oxychloride yields 92. Reaction of chloropyrimidine 92 with hydrazine produces hydrazinopyrimidine 93, which is then cyclized in the presence of ethyl orthoformate to give the bicyclic compound 94. Subsequent heating produces the Dimroth type rearrangement product bemitradine (95) [15]. 8. TRIAZOLOPYREDAZINES A derivative of an isomeric azapurine ring system interestingly exhibits bronchodilator activity, possibly indicating interaction with a target for theophylline. The starting pyridazine 97 is available from dichloro compound 96 by sequential replacement of the halogens. Treatment of 97 with formic acid supplies the missing carbon and cyclizes the intermediate formamide with consequent formation of zindotrine (98) [16].


169 [2CH2OEt OH

2CH2OEt

CO2Et (89)

(90); R = H (91); R = CHO

CH2OEt NHNH2

CH2OEt

(95) Me

(96)

(94) Me i.NHNH 2

(97)

(93) Me

(98)

9. PYRIMIDINOPYRAZINES Pyrimidinopyrazines related to folic acid have been investigated in some detail for their antimetabolic and antineoplastic activities. A related compound, which lacks one nitrogen atom, has been described as an antiproliferative agent, indicating it too has an effect on cell replication. Aldol condensation of the benzaldehyde 99 with ethyl acetoacetate gives the cinnamate 100. This is then reduced catalytically to the acetoacetate 101. Reaction of that keto ester with 2,4,6- triaminopyrimidine gives the product 102 which is subsequently chlorinated (103) and subjected to hydrogenolysis. There is thus formed piritrexim (104) [17]. 10. PYREDAZINODIAZEPINES The ready acceptance of angiotensin converting enzyme inhibitors as antihypertensive agents has, as noted previously in this volume, engendered intensive investigation into the limits of the


170 CO2Et

OMe

OMe

COMe OMe

OMe

(101)

(100)

NH2Me

NHoMe OMe

(104)

OMe (102); R-OH (103); R~C1

SAR in this series. A fused heterocycle ranks among the more unexpected nuclei for an ACE inhibitor. The fact that the end product interacts with a receptor site on an enzyme means that a single enantiomer will be responsible for the activity of the drug. The synthesis takes cognizance of this fact by choosing that enantiomer as the target; the precursors for both starting materials (105,106) thus consist of S enantiomers. The amino group further removed from the carboxyl in the precursor to 106 is sufficiently more reactive to afford the monobenzylation product on alkylation. Acylation of that intermediate with the mono acid chloride of the protected glutamic acid 105 leads to amide 107. Hydrogenolysis then serves to remove the benzyl protecting groups so as to afford the aminoacid 108. This is then cyclized to the bicyclic amide 110 via the acid chloride 109. Reduction of the diamide with diborane interestingly occurs selectively at the less hindered amide grouping to afford 111. The phthaloyl (Phth) protecting group is then removed in the usual way by reaction with hydrazine to afford 112. Construction of the remainder of the molecule consists in the conversion of the R enantiomer of the hydroxy acid 113, to its triflate derivative 114. This very reactive leaving group is readily displaced by the primary amino group in 112 with inversion of configuration; the three chiral centers in the product thus all have the S configuration. Brief exposure of the alkylation product to anhydrous strong acid serves to remove


171

the tert-butyl protecting group. There is thus obtained cilazapril (116) [18]. CO2CH2C6H5 HN

(105)

l CO21 -Bu (106)

Ph

CO2t-Bu

(107); R1 = OCH2C6H5; R2 = CH2C6H5 (1O9);R1 = C1;R2 = H

Phth = 0

(112)

(111) OSO2CF3 -CH2CH2CCO2Et H (114)

O CO2 Jt -Bu (HO) OH -CH2CH2CCO2Et (113)

-CH2CH2C— NH1 CO2R k (); iu (116); R » H 11. THIAZOLOPYRIMIDONES The involvement of serotonin (5-hydroxytryptamine) in disease states has been recognized for several decades. Research on antagonists awaited the recent development of methodology involving serotonin receptors. A thiazolopyrimidone serves as the nucleus for a pair of serotonin antagonists. The key intermediate 118 is in fact simply the lactonized form of 2-hydroxyethyl acetoacetate. Condensation of this p-keto ester can be visualized to involve initial attack on the reactive


172

butyrolactone by the ring nitrogen of thiazole 117; cyclodehydration of that hypothetical intermediate 119 gives the fused heterocycle 120. The terminal hydroxyl group is then converted to the corresponding chloride 121. Displacement of the halogen by piperidine 122 affords ritanserin (123) [19]; the analogous reaction with the related dihydrothiazolopyrimidone using piperidine 124 leads to setoperone (125) [20]. Me Me [2CH2OH (117)

(119)

(118)

NH

(122)

/\:H2CH2X (120); X « OH (121); X ^ C l

(123)

O HN

V - C —f

V-F

(124)

0

Dihydro 121 "CH2CH2N

) — C —

NOMe 1

Me y CO2H (38); X =•• C(C6H5)3 (39); X*= H 0

(37)1

Cefixime (47) is another cephalosporin antibiotic with an aminothiazole ring attached to an oximinoether containing side chain at C-7. An interesting and novel side chain is attached to C - 3 . T h e cephalosporins derived from naturally occurring 7-aminocephalosporanic acid have an acetyl function attached to the C-3 methylene. Esterases cleave this in the body and the resulting alcohols are less potent as antibiotics. O n e way of dealing with this undesirable metabolic transformation is to hydrogenolyze this moiety to a methyl function such as is seen with cefetamet above. In an apparent attempt to prepare homologues, transformation to an aldehyde followed by Wittig reactions produces a series of vinyl analogues with interesting properties in their own right.

(3-Lactam Antibiotics

185

One of these is cefixime. The reader will also note that the oximino ether moiety of cefixime is an acetic acid function rather than the methyl ether seen above. This substitution not only retains the p-lactamase resistance intended but also alters the isoelectric point of the drug allowing for more efficient absorption. This agent is found to have rather good activity against Gram-negative microorganisms, is resistant to many p-lactamases and is orally active. It does, however, have somewhat less potency against Gram-positives. This would apparently lodge it among the second generation cephalosporins. One of the syntheses of this agent starts with 7-aminocephalosporanic acid (40) which is hydrolyzed to the alcohol with base, converted to its Schiffs' base with salicylaldehyde, and esterified to its diphenylmethyl ester with diphenyldiazomethane to give protected intermediate 41. Reaction of this with phosphorous pentachloride and pyridine exchanges the hydroxyl function for a chlorine (42). The olefinic linkage of 43 is introduced by reaction with triphenylphosphine and Nal to produce the Wittig reagent; reaction with formaldehyde gives the methylene derivative. Careful acid treatment of 43 selectively removes the salicyl protecting group from C-7 and the desired side chain is put in place by reaction with blocked synthon 45 (prepared by activation of the carboxyl group by reaction with phosphoryl chloride and DMF [Vilsmeier reagent]). The resulting blocked C-3 vinyl analogue (46) is deprotected at the primary aromatic amino group by reaction with rnethanolic HC1 and the remaining protecting groups are removed by reaction with trifluoroacetic acid in anisole to complete this synthesis of cefixime (47) [12]. Other syntheses have been described for this agent and its analogues [13-151. Cefpimizole (51) appears to be less active in vitro than cefotaxime and cefoperazone and to have a somewhat narrower activity spectrum although some strains of Pseudomonas are susceptible. It is not orally active, but its performance in vivo appears superior to what would be expected from its in vitro data. Its synthesis begins by acylation of cephaloglycin (48) with the bis acid chloride of imidazole-4,5-dicarboxylic acid (49) to give amide 50. The acetyl moiety at C-3 of this intermediate is displaced with 4-pyridineethanesulfonic acid and sodium iodide to give cefpimazole (51) [16]. Just as with cefpimazole, cefepime (54) has a quaternary nitrogen substituent at C-3 whose electron withdrawing character and excellent nucleofugic properties increases the reactivity of the

P-Lactam Antibiotics

186 ••OH H2N. o

CO2H (40)

> y U o H

-

CO2CH(C6H5)2 (41)

•OH

CO2CH(C6H5)2 (42) NOCH2CO2CMc3

CO2CH(C6H5)2 (43) NOCH2CO2CMe3

(45)

-N. CO2CH(C6H5)2 (44)

(46) NOCH2CO2H

(47)

H N

H2N Me CO2H (48)

(49)

CO2H (50)

i v-coci N-4 COC1

CO2CH(C6H5)2

P-Lactam Antibiotics

187

C ' OzH CO2H (51) (3-lactam bond and also the antibacterial potency of the molecule. This derivative is prepared simply by displacement of the C-3 chloro moiety of suitably blocked cephalosporin 52 with Nmethylpyrrolidine to give intermediate 53 and this is deblocked with trifluoroacetic acid in the usual manner to give cefepime (54) [17]. NOMc

(54) Ceftiofur (57) differs from the preceding cephalosporin derivatives in that it has a thioester moiety at C-3. This can be introduced by displacement of the C-3 acetyl group of 7-aminocephalosporanic acid (40) with hydrogen sulfide and esterification with 2-furylcarboxylic acid to give synthon 56. This can in turn be reacted with trimethylsilylated oximinoether derivative 55 (itself obtained from the corresponding acid by reaction with dicyclohexylcarbodiimide and 1-hydroxybenzotriazole) to produce, after deprotecting, ceftiofur (57) [18], Cefmenoxime (61) is a third generation parenteral cephalosporin whose in vitro antimicrobial spectrum approximates that of cefotaxime. Its side chains consist of the common methyltetrazo lylthio group at C-3 and the familiar oximinoether aminothiazole moiety at C-7. It is synthesized

188

p-Lactam Antibiotics

(55)

(56) NOMe

CO2H (57) from acid chloride 58 and 7-aminocephalosporin synthon 59 to produce blocked amide 60. The utility of this blocking group is demonstrated by its ease of removal to give cefotaxime (61) by reaction with thiourea (displacement of chloride by sulfur and internal cyclization to remove the protecting group from the amine) [19]. NOMe it N

ClCH.CONH-^V (58)

„ VT H N

« \

s

_. . _ N-N

(59) OMe

CO2H (60); R = C1CH2CO (61); R - H Cefpiramide (64) is a third generation cephalosporin with a 1 -methyl-[lHJ-tetrazol-5-ylthiomethyl moiety at C-3 and an acylated r>hydroxyphenylglycine moiety at C-7. It includes in its activity spectrum reasonable potency in vitro against many strains of Pseudomonas. It can be synthesized in a variety of ways including condensation of cephalosporin antibiotic 63 with 6methyl-4-(l-H)-pyridone-3-carboxylic acid in the form of its active N-hydroxysuccinimide ester (62) to produce cefpiramide (64) [20,21]. Cefoperazone (66) is a third generation parenteral antipseudomonal cephalosporin containing an acylated C-7 side chain reminiscent of that of piperacillin. One of the simplest syntheses


189

n ivovles acyalotin of antb ioitci 65 wh ti 4~ethy-2l3,~doixo~-lpp ierazn iecarbonyl cholrd ie treithO yaH lmn ie to gvie cefoperazone (66) [22]. O H

y H H 2 ,|fc (6C 5O )2 Ifc (66) CO The chemcial structure of ceb fuperazone (70) is somewhat anaolgous to that however,the C7- sd ie chan i ured io funcotin is atached to an ap ilhatci am aromacit one and n i fact consits of the amnio acd i threonn ie. (3a-lctamase is enhanced further by n icorporaotin of a C7- meh toxylmoeiyt dervied conce cephamycnis. One of the syntheses of this agent starts wh ti reacotin of threo doixo-1-pp ierazn iecarbonyl cholrd ie and subsequent trm i ethyslyialtoin to gvie synh ton is condensed wth i protected 7a-mniocephaolsporanci acd i ester 68 by means of maet and N-ethym l orphon ile. Aetfr debolckn ig,the product (69) is converetd to anaolgue by protecotin wh ti ethyvln iyl ether and acd i and sequenatil reacotin d ie and t-butyh lypocholrtei folwed by debolckn ig n i acd i to produce ceb fupera


190 Nfe OH EtN

H2N

N-N i N

NCONH""

f

CO2CHPh2 (67)

(68)

Me /—\, ^ ElN NCONtT o

N~N 6

o^

(69)

CO2H

Me

EtN

N

MCONH'

(70)

—i

N N

CO2H

Cefotriaxone (73) has a 2,5-dihydro-6-hydroxy-2-rnethyl-3-thia-5-oxo-l,2,4-triazine moiety at C-3 in order to avoid s o m e of the side effects of cephalosporins attributed to the presence of an alkylthiatetrazolyl moiety at C-3 (antabuse-like acute alcohol intolerance and prolonged blood clotting times). In animal studies, cefbuperazone has been shown to be m o r e effective than cefmetazole and cefoperazone in infected mice. Of the various syntheses, one proceeds from 7aminocephalosporanic acid itself (40) by condensation with (formarnidothiazolyl)(rnethoxyimino)acetic acid (71) promoted by l,r-(carbonyldioxy)dibenzotriazole (formed by reaction of 1-hydroxybenzotriazole with trichloromethylchlorocarbonate-this converts the carboxylic acid moiety to the activated 1-hydroxybenzotriazole ester of 71) to give cephalosporin 72. This synthesis of ceftriaxone (73) concludes by displacement of the acetoxyl moiety of 72 with 2,5-dihydro-6-hydroxy-2-methyl-3-mercapto-5-oxo-l,2,4-triazine and sodium bicarbonate [24]. Cefmetazole (78) is a cephamycin-inspired cephalosporin differing from the mainstream c o m p o u n d s in having an aliphatic amide moiety attached to C-7. Its antibacterial spectrum is similar to the second generation agent cefoxitin. T h e synthesis starts with 7-aminocephalosporan-

191

p-Lactam Antibiotics NOMe

(40)

(71)

NOMe

CO2H (72) NOMe

HoN' CO2H Me

(73)

ic acid derivative 59 which is converted to its S c h i f f s base (75) with 3,5~di~t-butyl-4-hydroxybenzaldehyde (74). T h e imine (75) is oxidized to the iminoquinonemethide (76) with lead dioxide. Methanol is added 1,8 to this conjugated system to produce desired intermediate 77. T h e protecting S c h i f f s base is removed by exchange with Girard T reagent and the 7-aminocephalosporanic acid derivative so produced is acylated with cyanomethylthioacetic acid as its acid chloride to produce ceftnetazole (78) [25].

CHO

QMc N-N -i N CO2H (76)

(77)

9Me NCCH2SCH2CONH - 4 - f

N-N

CO2H (78)

I Me

1 CO2H

I

192


Cefotetan (84) stands out from the other cephalosporin antibiotics in that it possesses the unusual 1,3-dithietan side chain at C-7. This was introduced initially as the result of a molecular rearrangement and this interesting moiety was found to be consistent with excellent activity against many Gram-negative microorganisms (although not pseudomonads). Methods were developed for the deliberate incorporation of this functionality into molecules. The drug is rather similar to ceftazidime in its in vitro antimicrobial spectrum. Its synthesis begins with hydrolysis of 4-cyano-5-ethylthio-3-hydroxyisothiazole (79) with sodium hydroxide to produce acid 80. Reductive cleavage to thiol 81 was accomplished in good yield by reaction with sodium in liquid ammonia. This synthon was used to displace the bromo atom of cephalosporin antibiotic 82 to give antibiotic 83. Treatment of the latter with aqueous sodium bicarbonate leads to a facile rearrangement of the isothiazole, presumably by intramolecular displacement. The mechanism of this interesting reaction is not certain but may involve the process indicated by the arrows in formula 83. The product is in any case the antibiotic cefotetan (84) [26].

HO %

t

OMc 4

CO2H

N >-SH

(79);R = CN (80);R = CO2H

T CO2H (82)

(81)

CO2H

CO2H (83) 9Mc H?N< CO2H (84)

jjfe


193

4. MONOBACTAMS Utilizing a specially designed and intensive screen for novel P-lactam antibiotics from bacterial species, a series of monobactams (monocyclic p-lactam antibiotics) were ultimately discovered which not only possessed interesting anti-Gram-negative activity but possessed an intriguing Nsulfonic acid moiety attached to the nitrogen of the p-lactam ring. This group mimics the function of the carboxyl moiety of penicillins and cephalosporins. Previously it had also been believed that a fused strained bicyclic p-lactam system was necessary for significant antibacterial activity. While the natural monobactams do not appear to have clinically useful activity, a number of their totally synthetic analogues do. This field has been explored intensively in recent years and some of the fruits of this work are reported here. Aztreonam (93) is a wholly synthetic agent directed primarily against Gram-negative bacteria. The total chemical synthesis of aztreonam begins with N-t-butoxycarbonyl threonine-O-benzylhydroxylamide (85) which cyclizes to the monocyclic P-lactam intermediate (86) by use of an intramolecular Mitsunobu reaction with triphenylphosphine and diethylazodicarboxylate ("DEAD"). Careful hydrogenolysis is employed to remove the benzyl protecting group without cleaving the N-0 bond to give intermediate 87. The now superfluous N-hydroxyl group is removed by reduction with titanium trichloride to give azetidin-2-one 88. The t-BOC protecting group is removed by hydrolysis with trifluoroacetic acid and then replaced by a carbobenzyloxy moiety by reaction with benzyl chloroforrnate. The protected intermediate (89) is sulfonated with sulfur trioxide in DMF to give sulfonamide 90. The CBZ protecting group is removed by hydrogenolysis to produce amphoteric azetidinone 91. The desired acyl side chain is installed in the form of its diphenylmethyl ester using N-hydroxybenzotriazole as condensing agent to produce 92. Hydrolysis with trifluoroacetic acid leads finally to aztreonam (93) [27]. Carumonam (103) differs from aztreonam structurally in having an acetic acid function attached as an ether to the oximino moiety in the side chain and a carbamoylmethyl group in place of the methyl group at C-3. Like aztreonam it is primarily effective against Gram-negative bacteria. An interesting synthesis starts with a 2 + 2 cycloaddition reaction using carbobenzoxyglycine


194

CONHOCHC 2H 65 (85) if MeC CNkI . 3O (88) CH O NH. 6C 5H 2C O' (90) Iv^c M < •CO2CH(C6H5)2 N •\ SOH 3

S O R ((8867)R ;);=R CH=2C6H H5 CH O NH 6C 5H 2C

(89) (91) MeVlyie COH 2 N\SO3H

(92) (93) (94) and a complex imine prepared from methyl L-valinate (95). The requisite ketene deriva made in situ by reacting CBZ-glycine (94) with i-butyl chloroformate or i-propyl chloroforma The product of cycloadditon is primarily cis (96) and can be separated from the mixture by tional crystalization. Careful hydrolysis with potassium carbonate removes the w t o ester group (97) and then the carbamoyl moiety is introduced by reaction with dichlorophosphoryl isocyan folowed by reaction with aqueous bicarbonate to give 98. The chiral auxiliary now having its purpose can best be removed by anodic oxidation in acetic acid-triethylamine solution folow by hydrolysis with potassium carbonate to give partialy deblocked monobacatm 99. The rem der of the steps to carumonam consist of reaction with sulfur trioxide to the N-sulfonate (10 hydrogenolytic deblocking of the primary amine to amphoteric 101, additon of the protected s chain to give 102, and deblocking with trifluoroacetic acid to produce carumonam (103) [28 Other conceptualy interesting syntheses are available to the interested reader [29,30].


195

^.CH2OCONH ^

CO2H (95)

(94)

(96) .CH2OCONH2

(101)

(100)

NOCH2CO2H

NOCH2CO2CMc3

^JM^CH2OCONH2

:H2OCONH2 X

SO3H

(102)

J - \ O' (103)

A n u m b e r of m o n o b a c t a m s have been prepared in which the N-sulfonic acid moiety has been replaced by various other acidic groups. O n e means of closing the (3-lactam ring in this d r u g class invokes the normally suppressed nucleophilicity of the N-atom of the oxime moiety. This leaves an oxygen atom which is either replaced by a sulfonic acid group (as with a z t r e o n a m ) or utilized as a functional group to which a bioisosterically equivalent appendage, such as an acetic acid moiety is attached. This works in part because the electronic character of the bridging oxygen is somewhat analogous to that of the sulfur but also because the molecular size of the oxygen and the methylene carbon are roughly equivalent to that of sulfur so that the position in space of the acidic

196


moiety is roughly the same in aztreonam and in gloximonam (after deblocking to oximonam). In order to enhance the oral bioavailability of oximonam (104), a prodrug has been made by esterification of the carboxyl group with the t-butyl ester of hydroxyacetic acid (105). The product is prodrug gloximonam (106) [31]. Gloximonam is efficiently converted to oximonam in the body by metabolic processes. NOMc ,-Me O°4)

+

H(XH2CO2CMe3

^

X

rf U

OCH2CO2H

(105)

NOMc •v-C (106) The drug latentiated in the previous paragraph is the parenteral antibacterial monobactam oximonam (104). Its subclass of monobactams is sometimes referred to as the monosulfactams. As with the other monobactams, oximonam is primarily active against Gram-negative microorganisms. The synthesis begins with methyl N-carbobenzoxythreonine (107) which is converted to its oximinoyl amide with hydroxylamine then acetylated and finally cyclized to protected monobactam 108 under modified Mitsunobu conditions (triphenylphosphine and dimethyldiazodicarboxylate). Deacetylation with sodium carbonate produces 109, Ether formation, aided by a water soluble carbodiimide, with 2-trimethylsilylethylbromo acetate (110) leads to ether-ester 111. Hydrogenolytic deblocking of the primary amino group (hydrogen, Pd-C, HC1) and acylation leads as in previous examples to 112. Final deblocking to oximonam (113) is carried out with tetrabutylamrnoniurn fluoride [32]. OH ^CH2OCONH\I^Mc T~~\H

^

^>CH 2 OCONH X Afe I T y-f (108);R = COMe (109); R = H

+

BrCH2CO2CH2CH2SiMc3 (110)


197

(111) J° M e

NOMe H2N

(112)

rT

\)CH2CO2CH2CH2SiMc3

"V

" " >

(104)

REFERENCES 1. F. Sakamoto, S. Ikeda, R. Hirayama, M. Moriyama, M. Sotomura, and G. Tsukamoto, Chem. Pharm. Bull, 35, 642 (1987). 2. P. H. Bentley, J. P. Clayton, M. O. Boles, and R. J. Girven, /. Chem. Soc, Perkin Trans, 1, 2455 (1979). 3. J. B. A. Scalesciani, U. S. Patent 4,327,105 (1982) via Chem. Abstr., 97:72,191p (1982). 4. H. Tobiki, H. Yamada, T. Miyazaki, N. Tanno, H. Suzuki, K. Shimago, K. Okamura, and S. Ueda, Yakugaku Zasshi, 100,49 (1980). 5. D. G. Brenner and J. R. Knowles, Biochemistry, 23, 5833 (1984). 6. R. A. Volkmann, R, D. Carroll, R. B. Drolet, M. L. Elliot, and B. S. Moore, /. Org. Chem,, 47, 3344 (1982). 7. J. C. Kapur and H. P. Fasel, Tetrahedron Lett., 26, 3875 (1985). 8. I. Shinkai, R. A. Reamer, F. W. Hartner, T. Liu, and M. Sletzinger, Tetrahedron Lett., 23, 4903 (1982). 9. W. J. Leanza, K. J. Wildonger, T. W. Miller, and B. G. Christensen, /. Med. Chem., 22, 1435 (1979). 10. R. Scartazzini and K Bickel, Swiss, 605,990 (1978) via Chem. Abstr., 90: 54,959w (1978). 11. R. Boucourt, R. Heymes, A. Lutz, L. Penasse and J. Perronnet, Tetrahedron, 34, 2233 (1978). 12. H. Yamanaka, T. Chiba, K. Kawabata, H. Takasugi, T. Masugi, and T. Takaya, /. Antibiotics, 38,1738 (1985). 13. K. Kawabata, H. Yamanaka, H. Takasugi, and T. Takaya, /. Antibiotics, 39,404 (1986). 14. H. Yamanaka, K. Kawabata, K. Miyai, H. Takasugi, T. Kamimura, Y. Mine, and T. Takaya, /. Antibiotics, 39,101 (1986). 15. K. Kawabata, T. Masugi, and T. Takaya, /. Antibiotics, 39, 384 (1986).

198


16. Anon., Jpn. Kokai, Tokkyo Koho, 57/209,293 (1982) via Chem. Abstr., 98: 160,521 (1982). 17. T. Naito, S. Aburaki, H. Kamachi, Y. Narita, J. Okumura, and H. Kawaguchi, /. Antibiotics, 39,1092 (1986). 18. B. Labeeuw and A. Salhi, Eur. Pat. Appl, 36,812 (1981) via Chem. Abstr., 96: 68,712w (1981). 19. M. Ochiai, A. Morimoto, T. Miyawaki, Y. Matsushita, T. Okada, H. Natsugari, and M. Kida, J. Antibiotics, 39,171 (1981). 20. H. Yamada, H. Tobiki, K« Jimpo, K. Gooda, Y. Takeuchi, S. Ueda, T. Komatsu, T. Okuda, H. Noguchi, K. Irie, and T. Nakagome, J, Antibiotics, 36, 532 (1983). 21. H. Yamada, H. Tobiki, K. Jimpo, T. Komatsu, T. Okuda, H. Noguchi, and T. Nakagome, /. Antibiotics, 36, 543 (1983). 22. Anon., Jpn. Kokai Tokkyo Koho, 57/118,588 (1982) via Chem. Abstr., 99: 5,433x (1982). 23. S. Takano, I. Takadura, H. Ochiai, K. Momonoi, K. Tanaka, Y. Todo, T. Yasuda, M. Tai, Y. Fukuoka, H. Taki, and I. Saikawa, Yakugaku Zasshi, 102, 629 (1982). 24. W. J. Kim, C. H. Lee, B. J. Kim, and G. S. Lee, Brit. UK Pat. Appl 2,158,432 (1985) via Chem. Abstr., 105:152,813s (1985). 25. H, Nakao, H. Yanagisawa, B. Shimizu, M. Kaneko, M. Nagano, and S. Sugawara,./. Antibiotics, 2% 554 (1916). 26. M. Iwanami, T. Maeda, M. Fujimoto, Y. Nagano, N. Nagano, A. Yamazaki, TT. Shibanuma, K. Tamazawa, and K. Yano, Chem. Pharm. Bull., 28, 2629 (1980). 27. R. B. Sykes, W. L. Parker, C. M. Cimarusti, W. H. Koster, W. A. Slusarchyk, A. W. Fritz, and D. M. Floyd, Eur, Pat. Appl, 48,953 (1982) via Chem, Abstr., 97: 92,116w (1982). G. Loren, Drugs of the Future, 8, 301 (1983). 28. S. Hashiguchi, Y. Maeda, S. Kishimoto, and M. Ochiai, Heterocycles, 24, 2273 (1986). 29. M. Sendai, S. Hashiguchi, M. Tomimoto, S. Kishimoto, T. Matsuo, M. Kondo, and M. Ochiai, J. Antibiotics, 38, 346 (1985). 30. S. Kishimoto, M. Sendai, S. Hashiguchi, M. Tomimoto, Y. Satoh, T. Matsuo, M. Kondo, and M. Ochiai, /. Antibiotics, 36, 1421 (1983). 31. S. Kishimoto, M. Sendai, and M. Ochiai, Eur. Pat. Appl., 135,194 (1985) via Chem. Abstr., 104: 5,707m (1985). 32. S. R. Woulfe and M. J. Miller, /. Med. Chem., 28,1447 (1985).

11 Miscellaneous Heterocycles

The classification of medicinal agents in terms of their chemical structures has been a hallmark of this series. This system proves readily applicable for the majority of structural types. Rather complex structures which might strain such a scheme, such as for example, p-lactams, are often represented by a sufficient number of examples so that they can be collected in a special chapter. Some structures however, perhaps inevitably, do not readily fall into any neat category, hence this chapter on " Miscellaneous Heterocycles ." It is of passing interest that a number of the entries, such as, for example, the phenothiazines and benzodiazepines, belong to structural classes which merited chapters in their own right in previous volumes. The fact that the references to the preparation of these compounds tend to be somewhat old, suggest that they were in fact first prepared when those fields were under intensive investigation. The granting of an USAN to those agents in the much more recent period covered by this book (1982-1987) prdbably means that, for one reason or another, that they have been taken off the shelf for clinical development. 1. PHENOTHIAZINES The phenothiazine duoperone (5) combines structural elements found in phenothiazine and butyrophenone antipsychotic agents. Alkylation of substituted piperidine 1 with 3-chloropropanol affords the intermediate 2; treatment of this with thionyl chloride converts the terminal hydroxyl to chloride. Alkylation of the phenothiazine 4 with halide 3 affords the antipsychotic agent duoperone (5) [1]. A more polar phenothiazine derivative in which the side chain consists of an amino amide 199

200

Miscellaneous Heterocycles

exhibits the antiarrhythmic activity often found in such hindered amides. Acylation of phenothiazine 6 with 2-chloropropionyl chloride gives the haloamide 7; displacement of halogen with morpholine gives moricizine (8) [2].

(6)

rO 2. BENZOCYCLOHEPTAPYRIDINES Demethylation of the tricyclic antihistamine 9, with cyanogen bromide gives the secondary amine 10; acylation of that intermediate with ethyl chloroformate affords the nonsedating H-l antihistaminic agent loratidine (11) [3]. It is of interest that this compound does not contain the zwitterionic function which is thought to prevent passage through the blood-brain barrier, characteristic of this class of compounds.


201

3. CARBAZOLES Alkylation of the dimethylpiperazine 12 with 3-chloropropanol followed by treatment of the product 13 with thionyl chloride gives the halide 14. Alkylation of the anionfromcarbazole (15) with that halide leads to the tricyclic antipsychotic agent rimcazole (16) [4].

>Et (11)

(9) j R « Me (10); R= H HN NH

RCH2CH2CH2N NH (13);R = OH (14)R C1

Me

(15)

CH2CH2CH2N NH (16)

Me

4. DEBENZAZEPINES The imipramine analogue in which one of the methyl groups on the side chain nitrogen is replaced by a phenacyl group retains antidepressant activity. The starting material for this analogue, desipratnine (19), interestingly is an antidepressant drug in its ownright.Alkylation of 19 with gchlorophenacyl bromide (20) leads to lofepramine (21) [5],

202


CH2CH2CH2R (18); R = Q (19); R a NHMe

(17)

O CH2CH2CH2NCH2CMe (21) 5. DIBENZOXEPINES Two related tncyclic heterocyclic systems provide the nuclei for analgesic compounds which are not related structurally to either NS AIDs or opioids. Saponification of the nitrile group in 22 leads, via the corresponding acid, to acid chloride 23. Friedel-Crafts cyclization of 23 affords the ketone 24, Reaction of that ketone with 2-dimethylamino ethanethiol in the presence of boron trifluoride leads to formation of the enol thioether. Demethylation with phenylchloroformate gives fluradoline (26) [6]. Acylation of the dibenzoazaoxepin 27 with ethylchloroforrnate leads to the carbamate 28. Hydrazinolysis of the ester gives the hydrazide 29. Acylation of the remaining basic nitrogen on that intermediate with 5-chloropentanoic acid affords pinadoline (30) [7]. O

(23)

(26)

(24)


203

? f CNHNHCC (H C )1 42 N(27)

(28); R = Et (29); R = NHNH2

(30)

6. P Y R I D O B E N Z O D I A Z E P I N E S Antidepressant activity is retained w h e n the two carbon bridge in i m i p r a m i n e is replaced by a larger, m o r e complex, function. Nucleophilic aromatic substitution on chloropyridine 3 1 by means of o-aminobenzophenone (32) gives the bicyclic intermediate 3 3 . Reduction of the nitro group (34), followed by intramolecular Schiff base formation gives the required heterocyclic ring system 3 5 . Alkylation of the anion from 3 5 with l-dimethylamino-3-chloropropane leads to t a m p r a m i n e 36 [8].

N02 N

N—ci (32)

(31)

(33); R = O (34); R = H

H CH2CH2CH2NMe2 (35)

(36)

7. B E N Z O P Y R A N O P Y R I D I N E S T h e reaction of relatively simple starting materials, coumarin 4 0 , piperidone 3 7 and a m m o n i u m acetate, leads in a single step to the complex bridged tetracyclic c o m p o u n d 4 4 . T h e reaction can be rationalized by a s s u m i n g formation of the imine 3 8 from reaction of 3 7 , with ammonia. Conjugate addition of the eneamine-like tautomer 3 9 to the excellent Michael acceptor 4 0 will

204


lead to the adduct 41. This last is perfectly set up for intramolecular cyclization to the imide 42 by attack of imine nitrogen on the adjacent ester function. Ring opening of the lactone in 42 would afford the phenolic acid 43. This last intermediate is shown in neutral form; the reaction can be rationalized just as well by using the charged intermediate ions. Addition of the phenol ( or phenoxide) oxygen to the imide serves to close the last ring. Decarboxylation of the beta dicarbonyl grouping in 43 completes the reaction sequence. This last step can be visualized to occur either prior to or after the last ring is closed. The stereochemistry is probably thermodynamically controlled since the last ring closure in fact consists of carbinolamine formation from a ketone, a reaction which should be readily reversible. The product lortalamine (44) is described as an antidepressant agent [91.

NH

Me (38)

(44)

Me (39)

(43)


205

8. PYRROLOISOQUINOLINES Synthesis of the dopamine antagonist antipsychotic agent piquindone (53) starts by conversion of nicotinyl methyl ketone 45 to its acetal 46; alkylation with methyl iodide followed by reduction of the thus formed ternary iminium salt 47 by means of borohydride results in the tetrahydropyridine 48. Hydrolysis of the acetal group reveals the newly formed unsaturated ketone in 49. Conjugate addition of dimethyl malonate no doubt proceeds initially to the adduct 50; the reversible nature of the reaction as well as the ready enolization of the acetyl ring proton will result in formation of the more stable trans diequatorial isomer. The product cyclizes under the reaction conditions to perhydroisoquinolone 51. Hydrolysis of the ester in 51 is followed by decarboxylation to afford 52. Reaction of this last intermediate with 2-amino~3~pentanone under conditions for the Knorr reaction leads to formation of a pyrrole ring. The reaction sequence may be visualized by assuming initial imine formation between the quinolone and the aminoketone; cyclodehydration would complete the formation of the new ring. It is of note that a small amount of the isomeric angularly annulated product is formed as well. There is thus obtained piquindone (53) as the major reaction product [10]. 9. PYRAZOLOQUINOLINES Reaction of ethyl formate with perhydroisoquinolone 54, in the presence of strong base leads to the p-formylketone 55; condensation of that product with hydrazine leads to the formation of a new pyrazole ring. This product, quinpirole (56), is an antihypertensive agent [11], 10. NAPHTHOPYRANS The discovery of the utility of the bis-chromone carboxylic acid derivative cromolyn sodium in the treatment of asthma and related allergies has led to an intensive, and thus far not very fruitful, effort to discover analogues which would show oral activity in contrast to the lead which must be administered by inhalation. Preparation of a typical analogue, proxicromil (63), starts with the Oallylated phenol 57. Claisen rearrangement leads to the corresponding C-allylated product 58. The double bond is then reduced by catalytic hydrogenation (59). Base-catalyzed condensation


206 N V><s

o(46)

(45)

(47)

(48)

CO2Me CO2Me (51)

CO2Me J (50)

(49)

Mc\, Me

(52)

(53)

,N |

I n-CH2CH2CH3 (55) (54) (56) with diethyl oxalate can be envisaged as proceeding initially to the product from acylation of the n-CH 2 CH 2 CH 3

Q-CH2CH2CH3

ketone methyl group (60). Cyclodehydration of that product, shown in the enol form, will result in the formation of a pyran ring. There is thus obtained the chromone 61. Nitration of 61 produces nitro-pyran 62. The nitro group in 62 serves as a tool to introduce the phenolic hydroxyl group in 63 via reduction of the nitro group, diazotization of the resulting amino group, and treatment of the diazonium salt with aqueous sulfuric acid. During that treatment the ester functionality is hydrolyzed as well. There is thus obtained proxicromil (63) [12]. 11. BENZODIPYRANS In much the same vein, condensation of the difunctional o-hydroxyacetophenone 64 with diethyl

207


oxalate leads to the formation of two pyran rings on the same benzene ring. Workup under aqueous basic conditions leads to in situ saponification of the ester groups. There is thus obtained in a single step the mediator release inhibitor, probicromil (65) [13].

OCH2CH=CH2

CO2Et CH2CH2Me (61)

CH2CH2Me (60)

CO2H CH2CH2Me (63)

HO2C

CO2H CH2CH2Me (65) OMeO

+

MeSCH2COEt

IL.OH 'O*TCH2SMe

(67)

(68) OMeO

CH2SMe OMe (69)

J


208 12. FUROBENZOPYRANS

A closely related oxygenated heterocyclic system devoid of acidic groups interestingly shows quite different biological activity. Thus, condensation of the benzofuran hydroxyketone 66 with ethyl thiomethyl acetate (67) probably proceeds initially by formation of the acylation product 68. Intramolecular dehydration leads to formation of a pyran ring. There is thus obtained the hypocholesterolemic agent timefurone (69) [14]. 13. PYRANOQUINOLINES Mediator release inhibiting activity is, interestingly, retained when of one of the pyran rings in a molecule such as probicromil is replaced by a pyridine ring. The synthesis of such an agent starts with the allylation of hydroxyacetophenone 70 to give the ether 71. Rearrangement (72) followed by reduction gives the intermediate 73. The chromone ring is then added by base- catalyzed acylation with ethyl oxalate (74^. Removal of the acetyl group on nitrogen gives the intermediate, 75, required for construction of the pyridine ring. Thus, conjugate addition of the amino group in 75 to diethyl acetylene dicarboxylate gives the maleic ester 76. The ester function cyclizes onto the benzene ring on heating to form the pyridone 77. Saponification of the esters affords nedocromil (78) [15]. •N EC tOMc CH2R (70)

(72); R = CH=CH2 (73); R a Et

Et CHCO2Me %

CH2CH2Me (76)

R0 2 O MeCH2CH2 (77); R = Et (78); R = H

CO2Mc

Et02O

Et NR CH2CH2Mc (74); R = COMe (75);R = H


209

14. DIBENZOPYRANS Natural products have played an important role in the process of drug discovery. Several important medicinal agents were first identified as a result of the adventitious discovery of their biological activity. Familiar examples include the opioid analgesics, digitalis cardiotonic agents, and antihypertensive agents related to reserpine. The host of biological activities exhibited by marijuana suggested that this natural product should act as a lead to significant new classes of drugs. The identification and structural elucidation of the compound responsible for the biological activities, tetrahydrocannabinol (79), spurred interest in this compound. The development, in short order, of facile routes to this compound led to the synthesis of many new analogues. Acylation of the THC analogue 80, in which both the position of the double bond and the length of the side chain have been changed, with 4-(homopiperidino)butyric acid gives the ester nabazenil (82) which is described as an anticonvulsant agent. Esterification of the phenol in the isomeric THC-like compound 81 with 4-(diethylamino)butyric acid gives the ester 83, naboctate, in which antinauseant and antiglaucoma activity predominate [16].

OH

OH

(80); R = CHCH(CH2)4Me

(79)

MeMe Me

OCCH2CH2CH2NJ •CHCH(CH2)4Me MeMe

Me (82)

Me V J^ OCCH2CH2CH2NEt2 Me

| Me (83)


210 15. BENZOPYRANOPYRIDINES

The synthesis of THC analogues in which the carbocyclic ring is replaced by a heterocycle involves as a key step condensation of the resorcinol 84 with the appropriate (3-keto ester. Thus, condensation of that intermediate with piperidone 85, leads to the lactone 87. The reaction can be rationalized by assuming initial formation of the ester 86; the required carbon-carbon bond will then result from a cyclodehydration step. Reaction of the lactone with an excess of methyl Grignard reagent leads to the isolation of the THC analogue 88. This last reaction may in fact proceed initially to a tertiary carbinol of the general formula shown in 90. The tertiary carbocation formed under the acidic condition of the workup can in principle cyclize with the phenol to give the pyran ring of the product, 88, which is finally isolated. Debenzylation of 88 is followed by N-alkylation with propargyl bromide and acylation with 4,N-(l-methylpiperidino)-2-methylbutyric to afford the analgesic agent menabitan (89) [17]. 16. THIOPYRANOBENZOPYRANS The same sequence starting with the thiapyrone 91 affords via initially 92 the lactone 93. This intermediate gives the antihypertensive agent tinabinol (94) on reaction with excess methylmagnesium bromide (18).

(86)

CH2Ph

CH2Ph

(87)

(88)

CH 2 Ph

HOsaCCHo

OMe OCCHCH2CH2N~\

Me'

R

OH (90)

= CHCH(CH2)4Me Me Me

OH (91)


211

OH -l

OH

(92)

(93)

Me (94); R = CHCHCCH^Mc Me Me

17. PYRAZINOPYRIDOINDOLES Piperazine rings are a common structural feature in many compounds which act as antagonists at a-adrenergic receptors; it is of interest that a compound which incorporates a fused piperazine acts as an antihypertensive agent by virtue of a-blocking activity. Reaction of the indole 95 with ybutyrolactone leads to acid 96. Friedel-Crafts type cyclization affords the tricyclic ketone 97. Treatment of that intermediate with bromine gives the product from oc-bromination (98). Reaction of the bromoketone with ethylene diamine in the presence of borohydride leads to formation the fused piperazine ring(99). The sequence probably involves formation of the initial carbon-nitrogen bond by displacement of bromine; the second bond is then formed by reductive alkylation of the ketone. The steric environment around each of the two secondary amino groups is sufficiently different so that they show markedly different reactivities toward alkylation reactions. Thus, treatment with isopropyl bromide leads to the monoalkylation product 100. The remaining amino group is alkylated by reaction with ethyl bromide under more forcing conditions. There is thus produced atiprosin (101) [19].

•N H (95)

(96)

HR2 = H;R2 = (101);R1=Et;R2 =

(97); X = H (98); X = Br


212 18. THIENOBENZODIAZEPINES

Many examples of retention of activity in the face of replacement of benzene by thiophene have been noted so far. Application of this stratagem to a clozapine-like antipsychotic constitutes yet another example where activity is retained. The seven-membered ring of the compound in question is established by intramolecular amide formation on intermediate 102. Treatment of amide 103 with N-methylpiperazine in the presence of titanium tetrachloride affordsflumezapine(104) [20]. 19. IM1DAZOQUINAZOLINONES The growing awareness of the importance of chirality in biological activity has led to a trend to prepare new drugs in enantiomerically pure form. It is hoped by this means to spare the metabolic system the task of handling the inactive or even potentially toxic enantiomer. The more elegant solution to preparation of Such enantiomers involves the use of chiral synthons. Thus, alkylation of benzyl chloride 105 with ethyl D-alanine gives the chiral alkylation product 106, which is hydrogenated to produce the amine 107. Treatment with cyanogen bromide leads to the N-cyano intermediate 108. This undergoes a double ring closure under the reaction conditions to form the platelet aggregation inhibitor quazinone (109) [21]. ,NH2CO2Et

(102)

H (103) H2

93$ Cl M e (108) ÔEt

N

CH2C1

Cl (105)

O

(106); (107);

Cl

MeH (109)


213

20. IMIDAZOPURINES Imidazopyrimidines related to theophylline have been investigated extensively as a source for bronchodilating agents. Alkylation on nitrogen and fusion of an additional heterocyclic ring gives a bronchodilator which also displays some mediator release inhibiting activity. An additionelimination reaction of benzylamine 110 on imidazoline 111 leads to replacement of the thiomethyl group by the amino group (112). This intermediate is then treated in the same pot with nitroso cyanoacetamide (113). The reaction can be envisaged as involving addition of the benzylamine nitrogen to the cyano group and amide interchange between 113 and the imidazole nitrogen. The nitroso group is then reduced catalytically and the resulting diamine 115 cyclized by means of ethyl orthoformate and acetic anhydride to afford fenprinast (116) [22]. CH2NH2 MeS

(116) 21. PYRAZINOISOQUINOLINES One of the more interesting syntheses for the anthelmintic agent praziquantel (123) involves an Oppolzer type electrocyclization reaction. Reduction of the nitrile in benzocyclobutane 117 by means of LAH gives the corresponding amine 118. This is alkylated with chloroacet-


214

amide, and the product 119 is acylated with cyclohexylcarbonyl chloride. Reaction of the amide with formaldehyde in the presence of acetic anhydride leads to the carbinolamine acetate 121. Pyrolysis of this intermediate leads to ring opening of the benzocyclobutene to a bis-exomethylene cyclohexadiene and simultaneous loss of acetic acid from the carbinolamine acetate to form a methylene amine 122. Electrocyclic addition of the latter to the diene forms the required two rings at one time. There is thus obtained praziquantel (123) [23].

(117)

(118)

(119); R = H /—v (120); R = CO~\ ) \

f

-o (123) (122) 22. PYRAZ N IOPYRROLOBENZOD A IZEP N IES

(121)

Antidepressant activity is retained in a i n i a n s e r i n analogue in which a fused benzene ring is replaced by pyrrole e v e n though the t w o moieties differ in stereoelectronic properties. T h e starting nitrotoluene 124 is converted to the corresponding benzylamine 126 via bromide 125. Condensation of the a m i n e with the dimethoxy tetrahydrofuran 127 leads to formation of pyrrole 128; reduction of the nitro group then affords aniline 129. Reaction of that intermediate with the methyl hemiacetal from methyl glyoxylate leads to the tricyclic c o m p o u n d 130. This transformation may involve initial reaction of the aldehyde with the amino group; attack of the iminium derivative on the pyrrole will serve to close the ring. Acylation of the secondary amine with chloroacetyl chloride gives the amide 1 3 1 . T h e transient amine which results from displacement of chlorine b y methylamine reacts with the adjacent ester under the reaction conditions to form the r e m a i n i n g


215

ring of 132. Reduction of the amide functions by means of LAH leads to aptazapine (133) [24].

NR2

(124); R = H (125); R = Br (126); R = NH2

(128); R = 0 (129); R = H

R CO2Me (130); R = H (131); R = COCH2C1

u Me (133) 23. IMIDAZOQUINOLINES A pair of imidazoquinolines which differ only in substitution patterns differ markedly in biological activities. One of these agents, acodazole (139), shows antineoplastic activity while its seemingly close analogue, furodazole (143), is an anthelmintic agent. Synthesis of the former starts by reaction of 3,4-diaminonitrobenzene with formic acid. Reduction of the nitro group in the product 134 leads to the corresponding amine 135. Condensation of this last intermediate with ethyl acetoacetate proceeds initially to the imine 136; this cyclizes to quinoline 137 on heating. The hydroxyl group is then replaced by chlorine with phosphorus oxychloride (138). Displacement of halogen by means of N-methyl-N-acetyl-p,- phenylenediamine affords acodazole (139) [25]. Acylation of the common starting 3,4-diaminonitrobenzene with furoyl chloride proceeds on the more basic amino group meta to the nitro group to give 140. This is then cyclized to imidazole 141 by means of acetic anhydride. Reduction of the nitro group (142), followed by condensation with ethyl acetoacetate affords furodazole (143) [26]. 24. OXAZOLOQUINOLINES Nitration of quinoline 144 leads to the nitro derivative 145. Reduction of the nitro group leads to


216

the orf/io-aminophenol 146. Reaction of that intermediate with methyl trimethoxyacetate leads to formation of a benzoxazole ring. There is thus obtained the mediator release inhibitor agent quazolast (147) [27]. H

H2N H2N

S

NR2 •Me (136)

(134); R = O (135); R = H

NJLAN

H2N-

ccx,

^NO2

Me

(140)

(137);R = (138);R =

M^ (139)

(141); R a O (142); R = H

(143)

Cl

Cl

"NR, OH (144)

(145); R = O (146); R = H

CO2Me (147)


217

25. THIAZOLOBENZMIDAZOLES Thefindingthat the anthelmintic thiazoloimidazole levamisole showed immunoregulatory activity spurred further investigation of this heterocyclic system. Synthesis of a highly modified analogue starts by displacement of bromine in keto ester 149 by sulfur in substituted benzimidazole 148. Cyclization of the product (150), leads initially to the carbinol 151. Removal of the ester group by saponification in base followed by acid-catalyzed dehydration of the carbinol affords the immune regulator tilomisole (152) [28],

CH2CO2Et

CH2CO2Et

(151) 26. PYRIMIDOINDOLES The two carbonyl groups in isatin (153) show quite different reactivities since one is a ketone and the other an amide. Condensation of that compound with the Grignard reagent from o- chlorobromobenzene thus gives 154. Conjugate addition of the anion from that product to acrylonitrile affords the proprionitrile 155, The cyano group is then reduced to the primary amine by means of LAH. Internal imine formation leads to cyclization. There is thus obtained the antidepressant agent ciclazindol (157) [29].


218

27. ETHENOPYRROLOCYCLOBUTISOINDOLES The antineoplastic agent mitindomide (160) in fact represents the well-known product from irradiation of maleimide in benzene [30]. The activity of this old compound was uncovered by one of the large antitumor screens maintained by the National Cancer Institute. The structure is sufficiently complex and the starting materials sufficiently available to lead one to suspect that the product is still produced photochemically. The product can be rationalized by assuming successive 1,4 and 1,2 additions to benzene. Intermediate 159a involves the 1,4 followed by 1,2 addition; intermediate 159b presupposes the steps occur in the reverse order.

(153)

(154)

(155)

CH2CII2CN

CH2CH2CH2NH2 (157)

(156)

HH

The Organic Chemistry of Drug Synthesis

The Organic Chemistry of Drug Synthesis

The Organic Chemistry of Drug Synthesis